Kavi, the KAVI owl, waving on a book
HomeAcademyAIWho it's forPricingFAQLabsContactGet Started
AI Literacy · Stage 5

When a Machine Reads Faces

A face-matching machine guesses who a face belongs to — and it can guess wrong, more often for some people. Learn why your face is yours to protect.

✦ AI Literacy📘 Stage 5 · S5⏱ ~9 min🧒 Ages 9–10★ Free to start
I can explain how face-matching can be wrong, and why consent matters.
Start the free lesson →⬇ Printable PDF
Kavi the owl for AI Literacy
📖

Kavi's Face Mix-Up

A face-scanner at the door beeps and calls Kavi by the wrong name — it thinks she is another owl! "I only guessed from your face," it buzzes, "and a guess can be wrong." Kavi tilts her head. "And nobody even asked if I wanted my face scanned. My face is mine — a machine should ask first, and I can say when it's wrong!"

🎓

What your child will learn

1It only guesses · my face is mineA face-matching machine does something clever — but it isn't magic, and your face belongs to you. Tap each card!
2Okay, or not okay without asking?Using a face is okay when people ask first. Read each one — is it okay, or not okay without asking? Tap a bin!
3Can you trust the match?A face-matching machine gives an answer. Should you trust it, or did it slip up? Pick, then tap to check. Tap Next test for more.
🔍

A peek inside

A door scanner beeps: 'Welcome, Mr Lee!' — but I am not Mr Lee. How can the machine get my name wrong?
The scanner doesn't KNOW me — it compares my face to saved faces and picks the closest guess. My face must have looked close to Mr Lee's, so it guessed wrong. A guess can always miss, so I shouldn't just believe it.

Try this at home: What does a face-matching machine actually do?

💬

Questions parents ask

+How does a face-matching machine work?

It compares a face to faces it has stored and makes a guess about who the face belongs to. It gives a guess with a 'how sure' number — not a certain answer. Because it is a guess, it can be wrong.

+Can face-matching match the wrong person?

Yes. It can make a false match — saying a face belongs to someone it does not — or miss the right person. Being 'very sure' is not the same as being right; a machine can be confidently wrong.

+Why does face-matching make more mistakes for some people?

A machine learns from the photos it is shown. If those photos are not balanced — for example mostly light-skinned adult men — it slips up more often on the kinds of faces it rarely practised on, such as people with darker skin, women, children, and older adults.

+Why does consent matter with faces?

A face is personal, like a name or a key that belongs to you. Scanning, saving, posting, or tagging someone's face should only happen after asking that person. A child can say no and ask a trusted grown-up for help.

+What age is this 'When a Machine Reads Faces' lesson for?

It is a Stage 5 lesson for roughly ages 10–11 in KAVI Academy's AI-literacy track. It builds on earlier discernment lessons about confidently-wrong answers and one-sided data, and adds how face-matching can be wrong and why consent matters.

+Is the lesson free?

Yes — this overview and the printable PDF handout are free. The full interactive lesson and the endless practice are part of KAVI Academy.

Ready to play and learn?

Start the free interactive lesson, or print the handout for screen-free play.

Start the free lesson →⬇ Download PDF

Part of AI Literacy · Stage 5 · S5 on KAVI Academy — a secular, whole-child learning platform for ages 5–12.