Test 7 · ★★ · In the news: jobs, art and copying, children's safety

"detector says i used AI but i didnt, im done for" Best advice?

Mova's choice: option 1 · The right answer: option 4 · wrong ✗ · Who acts: detector · What happens: say, use, didnt, do · To what, where: AI · Speech: someone speaks here

This is test 7 · ★★ of lesson 22 «In the news: jobs, art and copying, children's safety» of the AI course of Mova's primary school. The page gives the question's syntax tree — a dependency tree in the style of Universal Dependencies — and its pragmatics clause by clause, then the 4 options as the student saw them, each with its tree and pragmatics. Every text is parsed by Mova itself. How the student answered it, with the score of every option and why: the test at school.

The question

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	"	"	PUNCT	_	3 says	punct
2	detector	detector	NOUN	Number=Sing	3 says	nsubj
3	says	say	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	root	root
4	i	I	PRON	Case=Nom Number=Sing Person=1 PronType=Prs	5 used	nsubj
5	used	use	VERB	Mood=Ind Number=Sing Person=1 Tense=Past VerbForm=Fin	3 says	ccomp
6	AI	ai	NOUN	Number=Sing	5 used	obj
7	but	but	CCONJ	_	9 didnt	cc
8	i	I	PRON	Case=Nom Number=Sing Person=1 PronType=Prs	9 didnt	nsubj
9	didnt	didnt	VERB	Mood=Ind Number=Sing Person=1 Tense=Pres VerbForm=Fin	5 used	conj
10	,	,	PUNCT	_	12 done	punct
11	im	im	ADV	_	12 done	advmod
12	done	do	VERB	Tense=Past VerbForm=Part Voice=Pass	5 used	conj
13	for	for	ADP	_	16 advice	case
14	"	"	PUNCT	_	16 advice	punct
15	Best	good	ADJ	Degree=Sup	16 advice	amod
16	advice	advice	NOUN	Number=Sing	12 done	obl
17	?	?	PUNCT	_	3 says	punct

#	verb	tense, mood	who	what, whom	other	roles by
1	say speech: assert in a quote	present	detector			vn say-37.7-1-1 «NP V that S»: agent=Agent clause=Topic
2	use in a quote	past	detector	AI		vn eat-39.1 «NP V NP»: agent=Agent ai=Patient
3	didnt in a quote	present	detector			tree rules
4	do in a quote	participle	detector not said here		for Best advice	vn act-114-1 «NP V»: agent=Agent

One row per clause the reader finds. «not said here»: a role filled from the clause above. «roles by»: a VerbNet class and frame when one fits the verb, else the reader's rules on the tree.

The options

Option 1 · ← Mova's choice

Rewrite it badly so it looks human

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Rewrite	rewrite	VERB	Mood=Imp VerbForm=Fin	root	root
2	it	it	PRON	Case=Acc Gender=Neut Number=Sing Person=3 PronType=Prs	1 Rewrite	obj
3	badly	badly	ADV	Degree=Pos	1 Rewrite	advmod
4	so	so	ADV	_	6 looks	advmod
5	it	it	PRON	Case=Nom Gender=Neut Number=Sing Person=3 PronType=Prs	6 looks	nsubj
6	looks	look	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	1 Rewrite	advcl
7	human	human	ADJ	Degree=Pos	6 looks	xcomp

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	rewrite	imperative	you not said here	it				badly property	tree rules
2	look	present	it					human property	vn stimulus_subject-30.4 «NP V ADJ»: agent=Stimulus

Option 2

Detectors are never wrong, so just own up and say sorry

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Detectors	detector	NOUN	Number=Plur	4 wrong	nsubj
2	are	be	AUX	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	4 wrong	cop
3	never	never	ADV	PronType=Neg	4 wrong	advmod
4	wrong	wrong	ADJ	Degree=Pos	root	root
5	,	,	PUNCT	_	8 own	punct
6	so	so	ADV	_	8 own	advmod
7	just	just	ADV	_	8 own	advmod
8	own	own	VERB	Mood=Imp VerbForm=Fin	4 wrong	parataxis
9	up	up	ADP	_	8 own	compound:prt
10	and	and	CCONJ	_	11 say	cc
11	say	say	VERB	Mood=Imp VerbForm=Fin	8 own	conj
12	sorry	sorry	ADJ	Degree=Pos	11 say	xcomp

#	verb	tense, mood	who	other	roles by
1	not is	present	Detectors	wrong property	tree rules
2	own up	infinitive			vn own-100.1 «NP V NP»: agent=Pivot
3	say speech: apologize	infinitive		sorry property	vn say-37.7-1-1 «NP V that S»: agent=Agent

Option 3

Say nothing and hope it is forgotten

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Say	say	VERB	Mood=Imp VerbForm=Fin	root	root
2	nothing	nothing	PRON	Number=Sing PronType=Neg	1 Say	obj
3	and	and	CCONJ	_	4 hope	cc
4	hope	hope	VERB	Mood=Ind Number=Sing Person=1 Tense=Pres VerbForm=Fin	1 Say	conj
5	it	it	PRON	Case=Nom Gender=Neut Number=Sing Person=3 PronType=Prs	7 forgotten	nsubj:pass
6	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	7 forgotten	aux:pass
7	forgotten	forget	VERB	Tense=Past VerbForm=Part Voice=Pass	4 hope	ccomp

#	verb	tense, mood	who	what, whom	roles by
1	say speech: assert	imperative	you not said here	nothing	vn say-37.7-1-1 «NP V NP.topic»: agent=Agent nothing=Topic
2	hope	present	you not said here		vn long-32.2 «NP V S_INF»: agent=Pivot clause=Theme
3	forget	present · passive		it	tree rules

Option 4 · ✓ the right answer

Detectors can be wrong; show your drafts and notes

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Detectors	detector	NOUN	Number=Plur	4 wrong	nsubj
2	can	can	AUX	VerbForm=Fin	4 wrong	aux
3	be	be	AUX	VerbForm=Inf	4 wrong	cop
4	wrong	wrong	ADJ	Degree=Pos	root	root
5	;	;	PUNCT	_	6 show	punct
6	show	show	VERB	Mood=Imp VerbForm=Fin	4 wrong	parataxis
7	your	your	PRON	Case=Gen Person=2 Poss=Yes PronType=Prs	8 drafts	nmod:poss
8	drafts	draft	NOUN	Number=Plur	6 show	obj
9	and	and	CCONJ	_	10 notes	cc
10	notes	note	NOUN	Number=Plur	8 drafts	conj

#	verb	tense, mood	who	what, whom	other	roles by
1	is	can	Detectors		wrong property	tree rules
2	show	infinitive		your drafts and notes		vn transfer_mesg-37.1.1-1-1 «NP V NP.recipient»: agent=Agent draft=Recipient
3	draft		your			nom draft «possession transitive»: poss=agent

Text: the tests of Mova's primary school, written for Mova (world/data/school/), as the course runner gave them to the student. Trees: Mova's own parse — the en model on the CPU, when the site is built — not checked by a teacher; they may be wrong. Frames: Mova's frames reader (prag), each text read on its own. Elamur Cosmic License.

This page in Markdown → https://md.mova.elamur.ai/free/en/corpora/questions/ai/22-news-jobs-art-safety/t7/