AIQ
AIQ
Part of the Deepfakes & Misinfo lesson guide. Teaching a different grade? 🌈 Explorer (5–7) · 💻 Hacker (11–14) · ⚡ Architect (15–18)
Start with a quick show of hands: "How many of you have seen a video online that turned out to be fake or edited?" Let a couple of students share examples — filters, pranks, edited clips. Then say:
"AI can create a video of anyone saying anything. A photo of an event that never happened. Welcome to the era of synthetic media."
Unpack "synthetic media" in plain terms: "synthetic" means made artificially, not grown or recorded naturally — so synthetic media is video, audio, or images a computer generated rather than a camera or microphone actually capturing. Ask: "If a computer can make a video of literally anyone saying literally anything, what problems could that cause?" Let students brainstorm for a minute before moving on — write a few of their ideas on the board, since you'll circle back to them at the end of the activity.
Open the lesson on the app and read the hook screen together, pointing out the four icons — 🎭 Deepfakes, 📰 Fake News, 🔍 Detection, 🛡️ Defense — as the roadmap for what you're about to cover: what these fakes are, why they're a problem, and how to catch them.
If a student pushes back with "but I can always tell when something's fake," that's worth addressing head-on before moving to the activity: "That might be true for the goofy, low-effort fakes you've seen so far. Today we're talking about a different level — deepfakes made by AI that are specifically designed to be hard to tell apart from the real thing, and that's exactly why we need more than just 'looking closely' as our only tool."
Work through the app's three scenes together, reading each fact aloud and pausing to add the context below — this age group can handle real examples, not just the bare fact. As you go, keep circling back to one running question: "Is this thing on the screen something a camera or microphone actually captured, or something a computer generated?" That single distinction is the thread running through all three scenes, and by the end students should be able to answer it instantly for each item the app shows them.
Face Swap: AI can put someone's face onto another person's body in video, nearly undetectably. Explain simply how this is possible: the AI has been shown a huge number of photos and video frames of a person's face from different angles, and it learns the patterns of that face well enough to generate new frames of it saying or doing something else. Voice Clone: AI can clone a voice from just a 3-second clip, and scammers use this to fake calls from family members. Pause here — this is the fact most worth dwelling on. Ask: "If someone called you and it sounded exactly like your mom or dad asking you to do something unusual, what would make you stop and double-check?" Real Photo: even real, uncropped photos can mislead through context — a real photo from a different event or year, presented as if it's from today, is technically "real" but still misleading. AI Images: show the class thispersondoesnotexist.com if you have a screen and internet access (preview it yourself first) — every refresh generates a brand-new, fully invented face of a person who has never existed. That's often the moment this topic really lands for this age group. It's worth being direct about the limits of the tour, too: "I'm showing you real invented faces and telling you about real fake videos and voices, but I'm not going to show you an actual deepfake video in class — most examples online aren't screened for a classroom, and it's safer to learn how they work than to go looking for one."
Fake News: AI can write a convincing fake news article, formatted to look exactly like real journalism, in seconds. Election Threats: a deepfake video of a politician saying something they never said, released right before an election, could genuinely mislead voters — ask the class why timing might matter here (a fake released right before people vote leaves little time for it to be debunked). Megaphone: contrast this with something that amplifies a REAL voice — it can still spread truth or lies, but at least the voice really belongs to the person speaking. AI Scams: tell the $243,000 story plainly — in a real 2019 case, criminals used an AI-cloned voice that sounded just like a company's CEO to trick an employee into wiring $243,000 to a fake account. Connect it back to the voice-clone fact from Scene 1: this wasn't science fiction, it already happened, using a technique this class just learned about. Ask the class: "If a fake voice can trick a trained employee at a real company out of a quarter million dollars, what chance would a random phone call have of tricking one of us?" — the point isn't to scare anyone, it's to establish that even careful, experienced adults have been fooled by this, so nobody should feel embarrassed about needing a check-first habit.
Look Closely: check for weird hand shapes, blurry edges, mismatched lighting, unnatural eye movements — but be honest that this is getting harder every year as the technology improves. Detection AI: some tools look for digital artifacts invisible to human eyes, like inconsistent pixel patterns. Source Check: is this from a verified account? Do other trusted sources report the same thing? When was it actually posted? Critical Thinking: read this one as the lesson's real takeaway — the best deepfake detector is a person's own brain, asking "does this seem too outrageous?" and "who benefits from me believing this?"
If time allows, run the app's practice round together as a class matching game: call out each item — AI face swap video, real photograph, AI voice cloning, megaphone, deepfake detector AI, visual inspection, AI-generated fake news, source checking — and have students vote thumbs-up (made by AI) or thumbs-down (not AI) before revealing the answer. Close by reading the app's summary together: "Deepfakes use AI to create fake videos, cloned voices, and fabricated images. The best defense combines AI detection with critical thinking!" Give the whole activity about 12–15 minutes.
A note on running this discussion: at this age, students can usually name the right instinct ("check with someone," "look for other sources") even if they haven't yet done it themselves in a real situation. Push gently for specifics — "check with someone" becomes stronger as "call the person back on a number I already know, not the number that called me" — since the specific version is the one that actually works when a scam call arrives for real.
Each question below has one wrong answer that sounds tempting because it repeats a word from the lesson without capturing the actual idea — it's worth reading all four choices out loud before students answer, rather than letting them jump at the first familiar-sounding phrase.
Close with: "We're living in a time where AI can generate video, photos, and voices that look and sound completely real. That doesn't mean you should stop trusting anything online — it means you check before you believe, the same way you'd check a rumor at school before repeating it." Revisit the list of problems the class brainstormed at the start and see how many of the real dangers you covered — scams, fake news, election manipulation — match what they predicted. Point out that the class predicted several of these dangers before ever hearing the real examples — the underlying risks of synthetic media aren't hard to imagine once you know the technology exists; the harder part is building the habit of checking before you're actually the one being fooled.
Extension activity (15–20 minutes): Run a "Spot the Fake" round. Prepare four or five short, one-sentence "headlines" ahead of time — some real (from a legitimate local or school news source) and some you've invented to sound just as plausible, mixing in a couple of Filipino context examples (a fake "text scam" alert, a fake viral local story). Read each headline aloud in the same neutral tone and have students vote real or fake, then explain their reasoning before you reveal the answer. Wrap up by asking: "What made the fake ones convincing? What would you actually need to do to check one of these for real?" — steering the answer back toward checking multiple sources, not just gut feeling.
If your class has extra time, add a role-play round: pair students up, and have one partner call the other on a pretend phone (two fingers to the ear works fine) claiming to be a family member who urgently needs a gift card code or a password "right now." Have the receiving partner practice saying out loud what they'd actually do — hang up and call the real person back on a known number, tell a parent immediately, never send anything before checking. Rotate so every student practices being the one who has to respond, not just the one hearing about the scam secondhand — rehearsing the actual words builds the habit far better than just discussing it in the abstract.