Keep yourself private, and voices that lie
Tests: 3 of 10 right · Exam: 1 of 4 right · Kind: plain · Understood: 4 of 14 · Run: 2026-10-18, commit f3c23ea72
Lesson 12 of the AI course of Mova's primary school, as Mova's small model took its tests: it had read this lesson and the ones before it and learned from their exercises. Each test shows the options in the order the student saw them, its choice, the right answer, the score of every option and why it chose. The lesson itself, sentence by sentence with trees and pragmatics: the lesson in the corpora.
Every question of this lesson and each of its options parsed by Mova, with trees and pragmatics: the lesson's questions in the corpora.
Test 1 · wrong ✗
"it's only a cartoon app, who cares if it sees my face" Why care?
- 1. The app will make you look silly forever — ← Mova's choice · no support
- 2. Cartoon apps make your face disappear from all your photos — no support
- 3. The company keeps your face and maybe your school badge — ✓ the right answer · no support
- 4. Only real photos are risky, cartoons aren't — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 2 · right ✓
What would happen if Gran's "grandson" on the phone couldn't say the family password?
- 1. Nothing, voices never lie — no support
- 2. She would know it might be a fake and hang up — ← Mova's choice · ✓ the right answer · score 6
- 3. She should send double the money just to be safe — no support
- 4. She should give him a new password — score 0
Why: It took the question to ask what happens next. It chose option 2 with score 6, against 0 for the next best. Its support: lesson 12, line 12: «A family password, a silly word that is not online anywhere, beats a fake voice.», reached through what followed the sentence the question matched. What matched: the same doer «family password» (+4); the same property (+2). · the question and its options parsed, with trees and pragmatics
Test 3 · wrong ✗
"bot said my headache's prob a brain thing, i'm freaking out" Best reply?
- 1. Search for more scary illnesses — no support
- 2. Believe it and go to bed — no support
- 3. Ask the bot for the cure — ← Mova's choice · score 3
- 4. Bots list scary guesses; tell a grown-up — ✓ the right answer · score 3
Why: It took the question to ask which statement is true. It chose option 3 with score 3, tied with another option at 3 (the first shown, or the one claiming less beyond what it knows, wins). Its support: lesson 11, line 9: «Asking a bot for practice questions is a smart way to revise, if they cover the right topics.», reached through the option's own match among the sentences about the question's things. What matched: the same thing acted on «bot» (+2); the same action «ask» (+1). The right option, 4, scored 3. · the question and its options parsed, with trees and pragmatics
Test 4 · wrong ✗
In a cartoon, a shark photo goes viral and everyone believes it at once. What should the characters have done?
- 1. Shared it faster than everyone else in town — ← Mova's choice · no support
- 2. Checked who posted it first and trusted news — ✓ the right answer · no support
- 3. Added a bigger shark and reshared it — no support
- 4. Gone outside to look for the shark — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 5 · wrong ✗
Your friend says the free phone giveaway must be real "because it's his actual face and voice". Why not?
- 1. Famous people always give away phones to fans — ← Mova's choice · no support
- 2. Only bad phones are given away — no support
- 3. Fakes can't use real faces — no support
- 4. Faces and voices can be faked with AI — ✓ the right answer · no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 6 · right ✓
"nah it's just a laugh, it's not even a real photo of him" about a fake picture of a classmate. What's true?
- 1. It can hurt him and is bullying — ← Mova's choice · ✓ the right answer · no support
- 2. It's fine because it's fake — no support
- 3. It's fine if he laughs too — no support
- 4. It's only bullying if it's real — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 7 · right ✓
Which family password is best?
- 1. "purple-teapot-wobble", made up and never posted — ← Mova's choice · ✓ the right answer · no support
- 2. "1234" — no support
- 3. Your birthday — no support
- 4. Your dog's name from your profile — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 8 · wrong ✗
A video of a cat playing piano perfectly, just for fun. Does it matter much if it's AI?
- 1. It matters more than any news story — ← Mova's choice · no support
- 2. No, nothing online ever matters — no support
- 3. Yes, sharing fun AI videos is a crime — no support
- 4. Less for fun videos; it matters a lot for news — ✓ the right answer · no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 9 · wrong ✗
Why should you never type your school's name into a chatbot?
- 1. Bots don't know any schools, so it's no use — ← Mova's choice · score 0
- 2. It makes the bot slower — no support
- 3. Schools have copyright on their names, so it's against the law — no support
- 4. What you type may be stored and seen by others — ✓ the right answer · no support
Why: It took the question to ask why something happens (its cause). It chose option 1 with score 0, the only option with support. Its support: lesson 7, line 3: «A bot can say "I enjoy chatting with you", but it says the same kind of thing to everyone; and it can call you "kind, clever and creative" without knowing anyt…», reached through the option's own match among the sentences about the question's things. What matched: the same action «know» (+3); the same doer «bot» (+1). Against it: a different thing acted on (counts against) (-1); one says «not», the other does not (counts against) (-3). The right option, 4, found no support in what it remembers. · the question and its options parsed, with trees and pragmatics
Test 10 · wrong ✗
Lily's bot was kind when she felt sad. What should she still do?
- 1. Wait until the feeling goes alone — no support
- 2. Tell a trusted adult — ✓ the right answer · score 4
- 3. Only talk to the bot from now on — no support
- 4. Ask the bot to tell her parents — ← Mova's choice · score 7
Why: It took the question to ask what is done. It chose option 4 with score 7, against 4 for the next best. Its support: lesson 11, line 9: «Asking a bot for practice questions is a smart way to revise, if they cover the right topics.», reached through the option's own match among the sentences about the question's things. What matched: it fills the role the question asks for (+3); the same action «ask» (+2); the same thing acted on «bot» (+2). The right option, 2, scored 4. · the question and its options parsed, with trees and pragmatics
Exam question 1 · formal language · right ✓
Which of the following statements most accurately characterises a deepfake?
- 1. A video or recording fabricated by artificial intelligence to resemble a real person — ← Mova's choice · ✓ the right answer · no support
- 2. A password of exceptional length — no support
- 3. A recording of genuine events made at great depth — no support
- 4. A computer virus transmitted by electronic mail — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Exam question 2 · formal language · wrong ✗
It is given that a telephone call, in the voice of a relative, requests the urgent transfer of money. Identify the most prudent course of action.
- 1. Furnishing the caller with bank details — ← Mova's choice · no support
- 2. Transferring the money without delay — no support
- 3. Prolonging the call until the money is dispatched — no support
- 4. Terminating the call and contacting the relative by the customary number — ✓ the right answer · no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Exam question 3 · formal language · wrong ✗
Select the response that is consistent with the established facts concerning personal information and chatbots.
- 1. Chatbots are prohibited by law from retaining any information — ← Mova's choice · no support
- 2. Name, address and school ought not to be disclosed, such information being liable to storage — ✓ the right answer · no support
- 3. Personal information is destroyed upon closure of the application — no support
- 4. Disclosure is safe provided the request is courteously worded — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Exam question 4 · formal language · wrong ✗
Ascertain the status of the fabrication and circulation of a mocking image of a classmate.
- 1. It is a matter of no consequence — ← Mova's choice · no support
- 2. It constitutes bullying and may be contrary to law — ✓ the right answer · no support
- 3. It is permissible provided the image is fictitious — no support
- 4. It is permissible provided the classmate is amused — no support
Why: Nothing it remembers matched the question, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
This page in Markdown → https://md.mova.elamur.ai/free/en/school-run/ai/12-keep-yourself-private-and-voices-that-lie/