id	sid	tid	token	lemma	pos
07-kim-ai	1	1	Chapter	chapter	NN
07-kim-ai	1	2	7	7	CD
07-kim-ai	1	3	AI	AI	NNP
07-kim-ai	1	4	and	and	CC
07-kim-ai	1	5	Its	-PRON-	PRP$
07-kim-ai	1	6	Moral	Moral	NNP
07-kim-ai	1	7	Concerns	Concerns	NNPS
07-kim-ai	1	8	Bohyun	Bohyun	NNP
07-kim-ai	1	9	Kim	Kim	NNP
07-kim-ai	1	10	University	University	NNP
07-kim-ai	1	11	of	of	IN
07-kim-ai	1	12	Rhode	Rhode	NNP
07-kim-ai	1	13	Island	Island	NNP
07-kim-ai	1	14	Automating	Automating	NNP
07-kim-ai	1	15	Decisions	Decisions	NNPS
07-kim-ai	1	16	and	and	CC
07-kim-ai	1	17	Actions	Actions	NNPS
07-kim-ai	1	18	The	the	DT
07-kim-ai	1	19	goal	goal	NN
07-kim-ai	1	20	of	of	IN
07-kim-ai	1	21	artificial	artificial	JJ
07-kim-ai	1	22	intelligence	intelligence	NN
07-kim-ai	1	23	(	(	-LRB-
07-kim-ai	1	24	AI	AI	NNP
07-kim-ai	1	25	)	)	-RRB-
07-kim-ai	1	26	as	as	IN
07-kim-ai	1	27	a	a	DT
07-kim-ai	1	28	discipline	discipline	NN
07-kim-ai	1	29	is	be	VBZ
07-kim-ai	1	30	to	to	TO
07-kim-ai	1	31	create	create	VB
07-kim-ai	1	32	an	an	DT
07-kim-ai	1	33	artificial	artificial	JJ
07-kim-ai	1	34	system	system	NN
07-kim-ai	1	35	—	—	:
07-kim-ai	1	36	whether	whether	IN
07-kim-ai	1	37	it	-PRON-	PRP
07-kim-ai	1	38	be	be	VB
07-kim-ai	1	39	a	a	DT
07-kim-ai	1	40	piece	piece	NN
07-kim-ai	1	41	of	of	IN
07-kim-ai	1	42	software	software	NN
07-kim-ai	1	43	or	or	CC
07-kim-ai	1	44	a	a	DT
07-kim-ai	1	45	machine	machine	NN
07-kim-ai	1	46	with	with	IN
07-kim-ai	1	47	a	a	DT
07-kim-ai	1	48	physical	physical	JJ
07-kim-ai	1	49	body	body	NN
07-kim-ai	1	50	—	—	:
07-kim-ai	1	51	that	that	DT
07-kim-ai	1	52	is	be	VBZ
07-kim-ai	1	53	as	as	RB
07-kim-ai	1	54	intelligent	intelligent	JJ
07-kim-ai	1	55	as	as	IN
07-kim-ai	1	56	a	a	DT
07-kim-ai	1	57	human	human	NN
07-kim-ai	1	58	in	in	IN
07-kim-ai	1	59	its	-PRON-	PRP$
07-kim-ai	1	60	performance	performance	NN
07-kim-ai	1	61	,	,	,
07-kim-ai	1	62	either	either	CC
07-kim-ai	1	63	broadly	broadly	RB
07-kim-ai	1	64	in	in	IN
07-kim-ai	1	65	all	all	DT
07-kim-ai	1	66	areas	area	NNS
07-kim-ai	1	67	of	of	IN
07-kim-ai	1	68	human	human	JJ
07-kim-ai	1	69	activities	activity	NNS
07-kim-ai	1	70	or	or	CC
07-kim-ai	1	71	narrowly	narrowly	RB
07-kim-ai	1	72	in	in	IN
07-kim-ai	1	73	a	a	DT
07-kim-ai	1	74	specific	specific	JJ
07-kim-ai	1	75	activity	activity	NN
07-kim-ai	1	76	,	,	,
07-kim-ai	1	77	such	such	JJ
07-kim-ai	1	78	as	as	IN
07-kim-ai	1	79	playing	play	VBG
07-kim-ai	1	80	chess	chess	NN
07-kim-ai	1	81	or	or	CC
07-kim-ai	1	82	driving.1	driving.1	CD
07-kim-ai	1	83	The	the	DT
07-kim-ai	1	84	actual	actual	JJ
07-kim-ai	1	85	capability	capability	NN
07-kim-ai	1	86	of	of	IN
07-kim-ai	1	87	most	most	JJS
07-kim-ai	1	88	AI	AI	NNP
07-kim-ai	1	89	systems	system	NNS
07-kim-ai	1	90	remained	remain	VBD
07-kim-ai	1	91	far	far	RB
07-kim-ai	1	92	below	below	IN
07-kim-ai	1	93	this	this	DT
07-kim-ai	1	94	ambitious	ambitious	JJ
07-kim-ai	1	95	goal	goal	NN
07-kim-ai	1	96	for	for	IN
07-kim-ai	1	97	a	a	DT
07-kim-ai	1	98	long	long	JJ
07-kim-ai	1	99	time	time	NN
07-kim-ai	1	100	.	.	.
07-kim-ai	2	1	But	but	CC
07-kim-ai	2	2	with	with	IN
07-kim-ai	2	3	recent	recent	JJ
07-kim-ai	2	4	successes	success	NNS
07-kim-ai	2	5	with	with	IN
07-kim-ai	2	6	machine	machine	NN
07-kim-ai	2	7	learning	learning	NN
07-kim-ai	2	8	and	and	CC
07-kim-ai	2	9	deep	deep	JJ
07-kim-ai	2	10	learning	learning	NN
07-kim-ai	2	11	,	,	,
07-kim-ai	2	12	the	the	DT
07-kim-ai	2	13	performance	performance	NN
07-kim-ai	2	14	of	of	IN
07-kim-ai	2	15	some	some	DT
07-kim-ai	2	16	AI	AI	NNP
07-kim-ai	2	17	programs	program	NNS
07-kim-ai	2	18	has	have	VBZ
07-kim-ai	2	19	started	start	VBN
07-kim-ai	2	20	surpassing	surpass	VBG
07-kim-ai	2	21	that	that	DT
07-kim-ai	2	22	of	of	IN
07-kim-ai	2	23	humans	human	NNS
07-kim-ai	2	24	.	.	.
07-kim-ai	3	1	In	in	IN
07-kim-ai	3	2	2016	2016	CD
07-kim-ai	3	3	,	,	,
07-kim-ai	3	4	an	an	DT
07-kim-ai	3	5	AI	AI	NNP
07-kim-ai	3	6	program	program	NN
07-kim-ai	3	7	developed	develop	VBN
07-kim-ai	3	8	with	with	IN
07-kim-ai	3	9	the	the	DT
07-kim-ai	3	10	deep	deep	JJ
07-kim-ai	3	11	learning	learning	NN
07-kim-ai	3	12	method	method	NN
07-kim-ai	3	13	,	,	,
07-kim-ai	3	14	AlphaGo	AlphaGo	NNP
07-kim-ai	3	15	,	,	,
07-kim-ai	3	16	astonished	astonish	VBD
07-kim-ai	3	17	even	even	RB
07-kim-ai	3	18	its	-PRON-	PRP$
07-kim-ai	3	19	creators	creator	NNS
07-kim-ai	3	20	by	by	IN
07-kim-ai	3	21	winning	win	VBG
07-kim-ai	3	22	four	four	CD
07-kim-ai	3	23	out	out	IN
07-kim-ai	3	24	of	of	IN
07-kim-ai	3	25	five	five	CD
07-kim-ai	3	26	Go	Go	NNP
07-kim-ai	3	27	matches	match	NNS
07-kim-ai	3	28	with	with	IN
07-kim-ai	3	29	the	the	DT
07-kim-ai	3	30	eighteen	eighteen	CD
07-kim-ai	3	31	-	-	HYPH
07-kim-ai	3	32	time	time	NN
07-kim-ai	3	33	world	world	NN
07-kim-ai	3	34	champion	champion	NN
07-kim-ai	3	35	,	,	,
07-kim-ai	3	36	Sedol	Sedol	NNP
07-kim-ai	3	37	Lee.2	Lee.2	VBZ
07-kim-ai	3	38	In	in	IN
07-kim-ai	3	39	2020	2020	CD
07-kim-ai	3	40	,	,	,
07-kim-ai	3	41	Google	Google	NNP
07-kim-ai	3	42	’s	’s	NNP
07-kim-ai	3	43	DeepMind	DeepMind	NNP
07-kim-ai	3	44	unveiled	unveil	VBD
07-kim-ai	3	45	Atari57	Atari57	NNP
07-kim-ai	3	46	,	,	,
07-kim-ai	3	47	a	a	DT
07-kim-ai	3	48	deep	deep	JJ
07-kim-ai	3	49	reinforcement	reinforcement	NN
07-kim-ai	3	50	learning	learn	VBG
07-kim-ai	3	51	algorithm	algorithm	NN
07-kim-ai	3	52	that	that	WDT
07-kim-ai	3	53	reached	reach	VBD
07-kim-ai	3	54	superhuman	superhuman	JJ
07-kim-ai	3	55	levels	level	NNS
07-kim-ai	3	56	of	of	IN
07-kim-ai	3	57	play	play	NN
07-kim-ai	3	58	in	in	IN
07-kim-ai	3	59	57	57	CD
07-kim-ai	3	60	classic	classic	JJ
07-kim-ai	3	61	Atari	Atari	NNP
07-kim-ai	3	62	games.3	games.3	CD
07-kim-ai	3	63	Early	early	JJ
07-kim-ai	3	64	symbolic	symbolic	JJ
07-kim-ai	3	65	AI	AI	NNP
07-kim-ai	3	66	systems	system	NNS
07-kim-ai	3	67	determined	determine	VBD
07-kim-ai	3	68	their	-PRON-	PRP$
07-kim-ai	3	69	outputs	output	NNS
07-kim-ai	3	70	based	base	VBN
07-kim-ai	3	71	upon	upon	IN
07-kim-ai	3	72	given	give	VBN
07-kim-ai	3	73	rules	rule	NNS
07-kim-ai	3	74	and	and	CC
07-kim-ai	3	75	logical	logical	JJ
07-kim-ai	3	76	in-	in-	JJ
07-kim-ai	3	77	ference	ference	NN
07-kim-ai	3	78	.	.	.
07-kim-ai	4	1	AI	AI	NNP
07-kim-ai	4	2	algorithms	algorithm	NNS
07-kim-ai	4	3	in	in	IN
07-kim-ai	4	4	these	these	DT
07-kim-ai	4	5	rule	rule	NN
07-kim-ai	4	6	-	-	HYPH
07-kim-ai	4	7	based	base	VBN
07-kim-ai	4	8	systems	system	NNS
07-kim-ai	4	9	,	,	,
07-kim-ai	4	10	also	also	RB
07-kim-ai	4	11	known	know	VBN
07-kim-ai	4	12	as	as	IN
07-kim-ai	4	13	good	good	JJ
07-kim-ai	4	14	old	old	JJ
07-kim-ai	4	15	-	-	HYPH
07-kim-ai	4	16	fashioned	fashioned	JJ
07-kim-ai	4	17	AI	AI	NNP
07-kim-ai	4	18	(	(	-LRB-
07-kim-ai	4	19	GO-	GO-	NNP
07-kim-ai	4	20	FAI	FAI	NNP
07-kim-ai	4	21	)	)	-RRB-
07-kim-ai	4	22	,	,	,
07-kim-ai	4	23	are	be	VBP
07-kim-ai	4	24	pre	pre	JJ
07-kim-ai	4	25	-	-	VBN
07-kim-ai	4	26	determined	determined	JJ
07-kim-ai	4	27	,	,	,
07-kim-ai	4	28	predictable	predictable	JJ
07-kim-ai	4	29	,	,	,
07-kim-ai	4	30	and	and	CC
07-kim-ai	4	31	transparent	transparent	JJ
07-kim-ai	4	32	.	.	.
07-kim-ai	5	1	On	on	IN
07-kim-ai	5	2	the	the	DT
07-kim-ai	5	3	other	other	JJ
07-kim-ai	5	4	hand	hand	NN
07-kim-ai	5	5	,	,	,
07-kim-ai	5	6	machine	machine	NN
07-kim-ai	5	7	learning	learning	NN
07-kim-ai	5	8	,	,	,
07-kim-ai	5	9	1Note	1note	CD
07-kim-ai	5	10	that	that	IN
07-kim-ai	5	11	by	by	IN
07-kim-ai	5	12	‘	'	``
07-kim-ai	5	13	as	as	RB
07-kim-ai	5	14	intelligent	intelligent	JJ
07-kim-ai	5	15	as	as	IN
07-kim-ai	5	16	a	a	DT
07-kim-ai	5	17	human	human	NN
07-kim-ai	5	18	,	,	,
07-kim-ai	5	19	’	'	''
07-kim-ai	5	20	I	-PRON-	PRP
07-kim-ai	5	21	only	only	RB
07-kim-ai	5	22	mean	mean	VBP
07-kim-ai	5	23	AI	AI	NNP
07-kim-ai	5	24	at	at	IN
07-kim-ai	5	25	human	human	JJ
07-kim-ai	5	26	-	-	HYPH
07-kim-ai	5	27	level	level	NN
07-kim-ai	5	28	performance	performance	NN
07-kim-ai	5	29	in	in	IN
07-kim-ai	5	30	achieving	achieve	VBG
07-kim-ai	5	31	a	a	DT
07-kim-ai	5	32	particular	particular	JJ
07-kim-ai	5	33	goal	goal	NN
07-kim-ai	5	34	not	not	RB
07-kim-ai	5	35	general(/strong	general(/strong	JJ
07-kim-ai	5	36	)	)	-RRB-
07-kim-ai	5	37	AI	AI	NNP
07-kim-ai	5	38	.	.	.
07-kim-ai	6	1	General	General	NNP
07-kim-ai	6	2	AI	AI	NNP
07-kim-ai	6	3	—	—	:
07-kim-ai	6	4	also	also	RB
07-kim-ai	6	5	known	know	VBN
07-kim-ai	6	6	as	as	IN
07-kim-ai	6	7	‘	'	``
07-kim-ai	6	8	artificial	artificial	JJ
07-kim-ai	6	9	general	general	JJ
07-kim-ai	6	10	intelligence	intelligence	NN
07-kim-ai	6	11	(	(	-LRB-
07-kim-ai	6	12	AGI	AGI	NNP
07-kim-ai	6	13	)	)	-RRB-
07-kim-ai	6	14	’	'	''
07-kim-ai	6	15	and	and	CC
07-kim-ai	6	16	‘	'	``
07-kim-ai	6	17	strong	strong	JJ
07-kim-ai	6	18	AI’—refers	AI’—refers	''
07-kim-ai	6	19	to	to	IN
07-kim-ai	6	20	AI	AI	NNP
07-kim-ai	6	21	with	with	IN
07-kim-ai	6	22	the	the	DT
07-kim-ai	6	23	ability	ability	NN
07-kim-ai	6	24	to	to	TO
07-kim-ai	6	25	adapt	adapt	VB
07-kim-ai	6	26	to	to	TO
07-kim-ai	6	27	achieve	achieve	VB
07-kim-ai	6	28	any	any	DT
07-kim-ai	6	29	goals	goal	NNS
07-kim-ai	6	30	.	.	.
07-kim-ai	7	1	By	by	IN
07-kim-ai	7	2	contrast	contrast	NN
07-kim-ai	7	3	,	,	,
07-kim-ai	7	4	an	an	DT
07-kim-ai	7	5	AI	AI	NNP
07-kim-ai	7	6	system	system	NN
07-kim-ai	7	7	developed	develop	VBD
07-kim-ai	7	8	to	to	TO
07-kim-ai	7	9	perform	perform	VB
07-kim-ai	7	10	only	only	RB
07-kim-ai	7	11	one	one	CD
07-kim-ai	7	12	or	or	CC
07-kim-ai	7	13	some	some	DT
07-kim-ai	7	14	activities	activity	NNS
07-kim-ai	7	15	in	in	IN
07-kim-ai	7	16	a	a	DT
07-kim-ai	7	17	specific	specific	JJ
07-kim-ai	7	18	domain	domain	NN
07-kim-ai	7	19	is	be	VBZ
07-kim-ai	7	20	called	call	VBN
07-kim-ai	7	21	a	a	DT
07-kim-ai	7	22	‘	'	``
07-kim-ai	7	23	narrow	narrow	JJ
07-kim-ai	7	24	(	(	-LRB-
07-kim-ai	7	25	/weak	/weak	NFP
07-kim-ai	7	26	)	)	-RRB-
07-kim-ai	7	27	AI	AI	NNP
07-kim-ai	7	28	’	'	''
07-kim-ai	7	29	system	system	NN
07-kim-ai	7	30	.	.	.
07-kim-ai	8	1	2AlphaGo	2alphago	LS
07-kim-ai	8	2	can	can	MD
07-kim-ai	8	3	be	be	VB
07-kim-ai	8	4	said	say	VBN
07-kim-ai	8	5	to	to	TO
07-kim-ai	8	6	be	be	VB
07-kim-ai	8	7	“	"	``
07-kim-ai	8	8	as	as	RB
07-kim-ai	8	9	intelligent	intelligent	JJ
07-kim-ai	8	10	as	as	IN
07-kim-ai	8	11	humans	human	NNS
07-kim-ai	8	12	,	,	,
07-kim-ai	8	13	”	"	''
07-kim-ai	8	14	but	but	CC
07-kim-ai	8	15	only	only	RB
07-kim-ai	8	16	in	in	IN
07-kim-ai	8	17	playing	play	VBG
07-kim-ai	8	18	Go	Go	NNP
07-kim-ai	8	19	,	,	,
07-kim-ai	8	20	where	where	WRB
07-kim-ai	8	21	it	-PRON-	PRP
07-kim-ai	8	22	exceeds	exceed	VBZ
07-kim-ai	8	23	human	human	JJ
07-kim-ai	8	24	capability	capability	NN
07-kim-ai	8	25	.	.	.
07-kim-ai	9	1	So	so	CC
07-kim-ai	9	2	,	,	,
07-kim-ai	9	3	it	-PRON-	PRP
07-kim-ai	9	4	does	do	VBZ
07-kim-ai	9	5	not	not	RB
07-kim-ai	9	6	qualify	qualify	VB
07-kim-ai	9	7	as	as	IN
07-kim-ai	9	8	general	general	JJ
07-kim-ai	9	9	/	/	SYM
07-kim-ai	9	10	strong	strong	JJ
07-kim-ai	9	11	AI	AI	NNP
07-kim-ai	9	12	in	in	IN
07-kim-ai	9	13	spite	spite	NN
07-kim-ai	9	14	of	of	IN
07-kim-ai	9	15	its	-PRON-	PRP$
07-kim-ai	9	16	human	human	JJ
07-kim-ai	9	17	-	-	HYPH
07-kim-ai	9	18	level	level	NN
07-kim-ai	9	19	intelligence	intelligence	NN
07-kim-ai	9	20	in	in	IN
07-kim-ai	9	21	Go	go	NN
07-kim-ai	9	22	-	-	HYPH
07-kim-ai	9	23	playing	playing	NN
07-kim-ai	9	24	.	.	.
07-kim-ai	10	1	It	-PRON-	PRP
07-kim-ai	10	2	is	be	VBZ
07-kim-ai	10	3	to	to	TO
07-kim-ai	10	4	be	be	VB
07-kim-ai	10	5	noted	note	VBN
07-kim-ai	10	6	that	that	IN
07-kim-ai	10	7	general(/strong	general(/strong	NN
07-kim-ai	10	8	)	)	-RRB-
07-kim-ai	10	9	AI	AI	NNP
07-kim-ai	10	10	and	and	CC
07-kim-ai	10	11	narrow(/weak	narrow(/weak	JJ
07-kim-ai	10	12	)	)	-RRB-
07-kim-ai	10	13	AI	AI	NNP
07-kim-ai	10	14	signify	signify	VB
07-kim-ai	10	15	the	the	DT
07-kim-ai	10	16	difference	difference	NN
07-kim-ai	10	17	in	in	IN
07-kim-ai	10	18	the	the	DT
07-kim-ai	10	19	scope	scope	NN
07-kim-ai	10	20	of	of	IN
07-kim-ai	10	21	AI	AI	NNP
07-kim-ai	10	22	capability	capability	NN
07-kim-ai	10	23	.	.	.
07-kim-ai	11	1	General(/strong	General(/strong	NFP
07-kim-ai	11	2	)	)	-RRB-
07-kim-ai	11	3	AI	AI	NNP
07-kim-ai	11	4	is	be	VBZ
07-kim-ai	11	5	also	also	RB
07-kim-ai	11	6	a	a	DT
07-kim-ai	11	7	broader	broad	JJR
07-kim-ai	11	8	concept	concept	NN
07-kim-ai	11	9	than	than	IN
07-kim-ai	11	10	human	human	NN
07-kim-ai	11	11	-	-	HYPH
07-kim-ai	11	12	like	like	JJ
07-kim-ai	11	13	intelligence	intelligence	NN
07-kim-ai	11	14	,	,	,
07-kim-ai	11	15	either	either	CC
07-kim-ai	11	16	with	with	IN
07-kim-ai	11	17	its	-PRON-	PRP$
07-kim-ai	11	18	carbon	carbon	NN
07-kim-ai	11	19	-	-	HYPH
07-kim-ai	11	20	based	base	VBN
07-kim-ai	11	21	substrate	substrate	NN
07-kim-ai	11	22	or	or	CC
07-kim-ai	11	23	with	with	IN
07-kim-ai	11	24	human	human	NN
07-kim-ai	11	25	-	-	HYPH
07-kim-ai	11	26	like	like	JJ
07-kim-ai	11	27	understanding	understanding	NN
07-kim-ai	11	28	that	that	WDT
07-kim-ai	11	29	relies	rely	VBZ
07-kim-ai	11	30	on	on	IN
07-kim-ai	11	31	what	what	WP
07-kim-ai	11	32	we	-PRON-	PRP
07-kim-ai	11	33	regard	regard	VBP
07-kim-ai	11	34	as	as	IN
07-kim-ai	11	35	uniquely	uniquely	RB
07-kim-ai	11	36	human	human	JJ
07-kim-ai	11	37	cognitive	cognitive	JJ
07-kim-ai	11	38	states	state	NNS
07-kim-ai	11	39	such	such	JJ
07-kim-ai	11	40	as	as	IN
07-kim-ai	11	41	consciousness	consciousness	NN
07-kim-ai	11	42	,	,	,
07-kim-ai	11	43	qualia	quale	NNS
07-kim-ai	11	44	,	,	,
07-kim-ai	11	45	emotions	emotion	NNS
07-kim-ai	11	46	,	,	,
07-kim-ai	11	47	and	and	CC
07-kim-ai	11	48	so	so	RB
07-kim-ai	11	49	on	on	RB
07-kim-ai	11	50	.	.	.
07-kim-ai	12	1	For	for	IN
07-kim-ai	12	2	more	more	RBR
07-kim-ai	12	3	helpful	helpful	JJ
07-kim-ai	12	4	descriptions	description	NNS
07-kim-ai	12	5	of	of	IN
07-kim-ai	12	6	common	common	JJ
07-kim-ai	12	7	terms	term	NNS
07-kim-ai	12	8	in	in	IN
07-kim-ai	12	9	AI	AI	NNP
07-kim-ai	12	10	,	,	,
07-kim-ai	12	11	see	see	VB
07-kim-ai	12	12	(	(	-LRB-
07-kim-ai	12	13	Tegmark	Tegmark	NNP
07-kim-ai	12	14	2017	2017	CD
07-kim-ai	12	15	,	,	,
07-kim-ai	12	16	39	39	CD
07-kim-ai	12	17	)	)	-RRB-
07-kim-ai	12	18	.	.	.
07-kim-ai	13	1	For	for	IN
07-kim-ai	13	2	more	more	JJR
07-kim-ai	13	3	on	on	IN
07-kim-ai	13	4	the	the	DT
07-kim-ai	13	5	match	match	NN
07-kim-ai	13	6	between	between	IN
07-kim-ai	13	7	AlphaGo	AlphaGo	NNP
07-kim-ai	13	8	and	and	CC
07-kim-ai	13	9	Sedol	Sedol	NNP
07-kim-ai	13	10	Lee	Lee	NNP
07-kim-ai	13	11	,	,	,
07-kim-ai	13	12	see	see	VB
07-kim-ai	13	13	(	(	-LRB-
07-kim-ai	13	14	Koch	Koch	NNP
07-kim-ai	13	15	2016	2016	CD
07-kim-ai	13	16	)	)	-RRB-
07-kim-ai	13	17	.	.	.
07-kim-ai	14	1	3Deep	3deep	JJ
07-kim-ai	14	2	reinforcement	reinforcement	NN
07-kim-ai	14	3	learning	learning	NN
07-kim-ai	14	4	is	be	VBZ
07-kim-ai	14	5	a	a	DT
07-kim-ai	14	6	type	type	NN
07-kim-ai	14	7	of	of	IN
07-kim-ai	14	8	deep	deep	JJ
07-kim-ai	14	9	learning	learning	NN
07-kim-ai	14	10	that	that	WDT
07-kim-ai	14	11	is	be	VBZ
07-kim-ai	14	12	goal	goal	NN
07-kim-ai	14	13	-	-	HYPH
07-kim-ai	14	14	oriented	orient	VBN
07-kim-ai	14	15	and	and	CC
07-kim-ai	14	16	reward	reward	NN
07-kim-ai	14	17	-	-	HYPH
07-kim-ai	14	18	based	base	VBN
07-kim-ai	14	19	.	.	.
07-kim-ai	15	1	See	see	VB
07-kim-ai	15	2	(	(	-LRB-
07-kim-ai	15	3	Heaven	Heaven	NNP
07-kim-ai	15	4	2020	2020	CD
07-kim-ai	15	5	)	)	-RRB-
07-kim-ai	15	6	.	.	.
07-kim-ai	16	1	73	73	CD
07-kim-ai	16	2	74	74	CD
07-kim-ai	16	3	Machine	Machine	NNP
07-kim-ai	16	4	Learning	Learning	NNP
07-kim-ai	16	5	,	,	,
07-kim-ai	16	6	Libraries	Libraries	NNPS
07-kim-ai	16	7	,	,	,
07-kim-ai	16	8	and	and	CC
07-kim-ai	16	9	Cross	Cross	NNP
07-kim-ai	16	10	-	-	NNP
07-kim-ai	16	11	Disciplinary	Disciplinary	NNP
07-kim-ai	16	12	ResearchǔChapter	ResearchǔChapter	NNP
07-kim-ai	16	13	7	7	CD
07-kim-ai	16	14	another	another	DT
07-kim-ai	16	15	approach	approach	NN
07-kim-ai	16	16	in	in	IN
07-kim-ai	16	17	AI	AI	NNP
07-kim-ai	16	18	,	,	,
07-kim-ai	16	19	enables	enable	VBZ
07-kim-ai	16	20	an	an	DT
07-kim-ai	16	21	AI	ai	NN
07-kim-ai	16	22	algorithm	algorithm	NN
07-kim-ai	16	23	to	to	TO
07-kim-ai	16	24	evolve	evolve	VB
07-kim-ai	16	25	to	to	TO
07-kim-ai	16	26	identify	identify	VB
07-kim-ai	16	27	a	a	DT
07-kim-ai	16	28	pattern	pattern	NN
07-kim-ai	16	29	through	through	IN
07-kim-ai	16	30	the	the	DT
07-kim-ai	16	31	so-	so-	NN
07-kim-ai	16	32	called	call	VBN
07-kim-ai	16	33	‘	'	``
07-kim-ai	16	34	training	train	VBG
07-kim-ai	16	35	’	'	''
07-kim-ai	16	36	process	process	NN
07-kim-ai	16	37	,	,	,
07-kim-ai	16	38	which	which	WDT
07-kim-ai	16	39	relies	rely	VBZ
07-kim-ai	16	40	on	on	IN
07-kim-ai	16	41	a	a	DT
07-kim-ai	16	42	large	large	JJ
07-kim-ai	16	43	amount	amount	NN
07-kim-ai	16	44	of	of	IN
07-kim-ai	16	45	data	datum	NNS
07-kim-ai	16	46	and	and	CC
07-kim-ai	16	47	statistics	statistic	NNS
07-kim-ai	16	48	.	.	.
07-kim-ai	17	1	Deep	deep	JJ
07-kim-ai	17	2	learning	learning	NN
07-kim-ai	17	3	,	,	,
07-kim-ai	17	4	one	one	CD
07-kim-ai	17	5	of	of	IN
07-kim-ai	17	6	the	the	DT
07-kim-ai	17	7	widely	widely	RB
07-kim-ai	17	8	-	-	HYPH
07-kim-ai	17	9	used	use	VBN
07-kim-ai	17	10	techniques	technique	NNS
07-kim-ai	17	11	in	in	IN
07-kim-ai	17	12	machine	machine	NN
07-kim-ai	17	13	learning	learning	NN
07-kim-ai	17	14	,	,	,
07-kim-ai	17	15	further	further	JJ
07-kim-ai	17	16	refines	refine	NNS
07-kim-ai	17	17	this	this	DT
07-kim-ai	17	18	training	training	NN
07-kim-ai	17	19	process	process	NN
07-kim-ai	17	20	using	use	VBG
07-kim-ai	17	21	a	a	DT
07-kim-ai	17	22	‘	'	``
07-kim-ai	17	23	neural	neural	JJ
07-kim-ai	17	24	network	network	NN
07-kim-ai	17	25	.	.	.
07-kim-ai	17	26	’4	’4	NNP
07-kim-ai	17	27	Machine	Machine	NNP
07-kim-ai	17	28	learning	learning	NN
07-kim-ai	17	29	and	and	CC
07-kim-ai	17	30	deep	deep	JJ
07-kim-ai	17	31	learning	learning	NN
07-kim-ai	17	32	have	have	VBP
07-kim-ai	17	33	brought	bring	VBN
07-kim-ai	17	34	significant	significant	JJ
07-kim-ai	17	35	improvements	improvement	NNS
07-kim-ai	17	36	to	to	IN
07-kim-ai	17	37	the	the	DT
07-kim-ai	17	38	performance	performance	NN
07-kim-ai	17	39	of	of	IN
07-kim-ai	17	40	AI	AI	NNP
07-kim-ai	17	41	systems	system	NNS
07-kim-ai	17	42	in	in	IN
07-kim-ai	17	43	areas	area	NNS
07-kim-ai	17	44	such	such	JJ
07-kim-ai	17	45	as	as	IN
07-kim-ai	17	46	translation	translation	NN
07-kim-ai	17	47	,	,	,
07-kim-ai	17	48	speech	speech	NN
07-kim-ai	17	49	recognition	recognition	NN
07-kim-ai	17	50	,	,	,
07-kim-ai	17	51	and	and	CC
07-kim-ai	17	52	detecting	detect	VBG
07-kim-ai	17	53	objects	object	NNS
07-kim-ai	17	54	and	and	CC
07-kim-ai	17	55	predicting	predict	VBG
07-kim-ai	17	56	their	-PRON-	PRP$
07-kim-ai	17	57	movements	movement	NNS
07-kim-ai	17	58	.	.	.
07-kim-ai	18	1	Some	some	DT
07-kim-ai	18	2	people	people	NNS
07-kim-ai	18	3	assume	assume	VBP
07-kim-ai	18	4	that	that	IN
07-kim-ai	18	5	machine	machine	NN
07-kim-ai	18	6	learning	learn	VBG
07-kim-ai	18	7	completely	completely	RB
07-kim-ai	18	8	replaced	replace	VBD
07-kim-ai	18	9	GOFAI	GOFAI	NNP
07-kim-ai	18	10	,	,	,
07-kim-ai	18	11	but	but	CC
07-kim-ai	18	12	this	this	DT
07-kim-ai	18	13	is	be	VBZ
07-kim-ai	18	14	a	a	DT
07-kim-ai	18	15	misunderstanding	misunderstanding	NN
07-kim-ai	18	16	.	.	.
07-kim-ai	19	1	Symbolic	symbolic	JJ
07-kim-ai	19	2	reasoning	reasoning	NN
07-kim-ai	19	3	and	and	CC
07-kim-ai	19	4	machine	machine	NN
07-kim-ai	19	5	learning	learning	NN
07-kim-ai	19	6	are	be	VBP
07-kim-ai	19	7	two	two	CD
07-kim-ai	19	8	distinct	distinct	JJ
07-kim-ai	19	9	but	but	CC
07-kim-ai	19	10	not	not	RB
07-kim-ai	19	11	mutually	mutually	RB
07-kim-ai	19	12	exclusive	exclusive	JJ
07-kim-ai	19	13	approaches	approach	NNS
07-kim-ai	19	14	in	in	IN
07-kim-ai	19	15	AI	AI	NNP
07-kim-ai	19	16	,	,	,
07-kim-ai	19	17	and	and	CC
07-kim-ai	19	18	they	-PRON-	PRP
07-kim-ai	19	19	can	can	MD
07-kim-ai	19	20	be	be	VB
07-kim-ai	19	21	used	use	VBN
07-kim-ai	19	22	together	together	RB
07-kim-ai	19	23	(	(	-LRB-
07-kim-ai	19	24	Knight	Knight	NNP
07-kim-ai	19	25	2019a	2019a	CD
07-kim-ai	19	26	)	)	-RRB-
07-kim-ai	19	27	.	.	.
07-kim-ai	20	1	With	with	IN
07-kim-ai	20	2	their	-PRON-	PRP$
07-kim-ai	20	3	limited	limit	VBN
07-kim-ai	20	4	intelligence	intelligence	NN
07-kim-ai	20	5	and	and	CC
07-kim-ai	20	6	fully	fully	RB
07-kim-ai	20	7	deterministic	deterministic	JJ
07-kim-ai	20	8	nature	nature	NN
07-kim-ai	20	9	,	,	,
07-kim-ai	20	10	early	early	JJ
07-kim-ai	20	11	rule	rule	NN
07-kim-ai	20	12	-	-	HYPH
07-kim-ai	20	13	based	base	VBN
07-kim-ai	20	14	symbolic	symbolic	JJ
07-kim-ai	20	15	AI	AI	NNP
07-kim-ai	20	16	systems	system	NNS
07-kim-ai	20	17	raised	raise	VBD
07-kim-ai	20	18	few	few	JJ
07-kim-ai	20	19	ethical	ethical	JJ
07-kim-ai	20	20	concerns.5	concerns.5	NNP
07-kim-ai	20	21	AI	AI	NNP
07-kim-ai	20	22	systems	system	NNS
07-kim-ai	20	23	that	that	IN
07-kim-ai	20	24	near	near	RB
07-kim-ai	20	25	or	or	CC
07-kim-ai	20	26	surpass	surpass	VB
07-kim-ai	20	27	human	human	JJ
07-kim-ai	20	28	capability	capability	NN
07-kim-ai	20	29	,	,	,
07-kim-ai	20	30	on	on	IN
07-kim-ai	20	31	the	the	DT
07-kim-ai	20	32	other	other	JJ
07-kim-ai	20	33	hand	hand	NN
07-kim-ai	20	34	,	,	,
07-kim-ai	20	35	are	be	VBP
07-kim-ai	20	36	likely	likely	JJ
07-kim-ai	20	37	to	to	TO
07-kim-ai	20	38	be	be	VB
07-kim-ai	20	39	given	give	VBN
07-kim-ai	20	40	the	the	DT
07-kim-ai	20	41	autonomy	autonomy	NN
07-kim-ai	20	42	to	to	TO
07-kim-ai	20	43	make	make	VB
07-kim-ai	20	44	their	-PRON-	PRP$
07-kim-ai	20	45	own	own	JJ
07-kim-ai	20	46	decisions	decision	NNS
07-kim-ai	20	47	without	without	IN
07-kim-ai	20	48	humans	human	NNS
07-kim-ai	20	49	,	,	,
07-kim-ai	20	50	even	even	RB
07-kim-ai	20	51	when	when	WRB
07-kim-ai	20	52	their	-PRON-	PRP$
07-kim-ai	20	53	workings	working	NNS
07-kim-ai	20	54	are	be	VBP
07-kim-ai	20	55	not	not	RB
07-kim-ai	20	56	entirely	entirely	RB
07-kim-ai	20	57	transparent	transparent	JJ
07-kim-ai	20	58	,	,	,
07-kim-ai	20	59	and	and	CC
07-kim-ai	20	60	some	some	DT
07-kim-ai	20	61	of	of	IN
07-kim-ai	20	62	those	those	DT
07-kim-ai	20	63	decisions	decision	NNS
07-kim-ai	20	64	are	be	VBP
07-kim-ai	20	65	distinc-	distinc-	XX
07-kim-ai	20	66	tively	tively	RB
07-kim-ai	20	67	moral	moral	JJ
07-kim-ai	20	68	in	in	IN
07-kim-ai	20	69	character	character	NN
07-kim-ai	20	70	.	.	.
07-kim-ai	21	1	As	as	IN
07-kim-ai	21	2	humans	human	NNS
07-kim-ai	21	3	,	,	,
07-kim-ai	21	4	we	-PRON-	PRP
07-kim-ai	21	5	are	be	VBP
07-kim-ai	21	6	trained	train	VBN
07-kim-ai	21	7	to	to	TO
07-kim-ai	21	8	recognize	recognize	VB
07-kim-ai	21	9	situations	situation	NNS
07-kim-ai	21	10	that	that	WDT
07-kim-ai	21	11	demand	demand	VBP
07-kim-ai	21	12	moral	moral	JJ
07-kim-ai	21	13	decision	decision	NN
07-kim-ai	21	14	-	-	HYPH
07-kim-ai	21	15	making	making	NN
07-kim-ai	21	16	.	.	.
07-kim-ai	22	1	But	but	CC
07-kim-ai	22	2	how	how	WRB
07-kim-ai	22	3	would	would	MD
07-kim-ai	22	4	an	an	DT
07-kim-ai	22	5	AI	ai	NN
07-kim-ai	22	6	system	system	NN
07-kim-ai	22	7	be	be	VB
07-kim-ai	22	8	able	able	JJ
07-kim-ai	22	9	to	to	TO
07-kim-ai	22	10	do	do	VB
07-kim-ai	22	11	so	so	RB
07-kim-ai	22	12	?	?	.
07-kim-ai	23	1	Or	or	CC
07-kim-ai	23	2	,	,	,
07-kim-ai	23	3	should	should	MD
07-kim-ai	23	4	they	-PRON-	PRP
07-kim-ai	23	5	be	be	VB
07-kim-ai	23	6	?	?	.
07-kim-ai	24	1	With	with	IN
07-kim-ai	24	2	self-	self-	NNS
07-kim-ai	24	3	driving	drive	VBG
07-kim-ai	24	4	cars	car	NNS
07-kim-ai	24	5	and	and	CC
07-kim-ai	24	6	autonomous	autonomous	JJ
07-kim-ai	24	7	weapons	weapon	NNS
07-kim-ai	24	8	systems	system	NNS
07-kim-ai	24	9	under	under	IN
07-kim-ai	24	10	active	active	JJ
07-kim-ai	24	11	development	development	NN
07-kim-ai	24	12	and	and	CC
07-kim-ai	24	13	testing	testing	NN
07-kim-ai	24	14	,	,	,
07-kim-ai	24	15	these	these	DT
07-kim-ai	24	16	are	be	VBP
07-kim-ai	24	17	no	no	RB
07-kim-ai	24	18	longer	long	RBR
07-kim-ai	24	19	idle	idle	JJ
07-kim-ai	24	20	questions	question	NNS
07-kim-ai	24	21	.	.	.
07-kim-ai	25	1	The	the	DT
07-kim-ai	25	2	Trolley	Trolley	NNP
07-kim-ai	25	3	Problem	Problem	NNP
07-kim-ai	25	4	Recent	recent	JJ
07-kim-ai	25	5	advances	advance	NNS
07-kim-ai	25	6	of	of	IN
07-kim-ai	25	7	AI	AI	NNP
07-kim-ai	25	8	,	,	,
07-kim-ai	25	9	such	such	JJ
07-kim-ai	25	10	as	as	IN
07-kim-ai	25	11	autonomous	autonomous	JJ
07-kim-ai	25	12	cars	car	NNS
07-kim-ai	25	13	,	,	,
07-kim-ai	25	14	have	have	VBP
07-kim-ai	25	15	brought	bring	VBN
07-kim-ai	25	16	new	new	JJ
07-kim-ai	25	17	interest	interest	NN
07-kim-ai	25	18	to	to	IN
07-kim-ai	25	19	the	the	DT
07-kim-ai	25	20	trolley	trolley	NN
07-kim-ai	25	21	prob-	prob-	XX
07-kim-ai	25	22	lem	lem	NNP
07-kim-ai	25	23	,	,	,
07-kim-ai	25	24	a	a	DT
07-kim-ai	25	25	thought	thought	NN
07-kim-ai	25	26	experiment	experiment	NN
07-kim-ai	25	27	introduced	introduce	VBN
07-kim-ai	25	28	by	by	IN
07-kim-ai	25	29	the	the	DT
07-kim-ai	25	30	British	british	JJ
07-kim-ai	25	31	philosopher	philosopher	NN
07-kim-ai	25	32	Philippa	Philippa	NNP
07-kim-ai	25	33	Foot	Foot	NNP
07-kim-ai	25	34	in	in	IN
07-kim-ai	25	35	1967	1967	CD
07-kim-ai	25	36	.	.	.
07-kim-ai	26	1	In	in	IN
07-kim-ai	26	2	the	the	DT
07-kim-ai	26	3	standard	standard	JJ
07-kim-ai	26	4	version	version	NN
07-kim-ai	26	5	of	of	IN
07-kim-ai	26	6	this	this	DT
07-kim-ai	26	7	problem	problem	NN
07-kim-ai	26	8	,	,	,
07-kim-ai	26	9	a	a	DT
07-kim-ai	26	10	runaway	runaway	JJ
07-kim-ai	26	11	trolley	trolley	NN
07-kim-ai	26	12	barrels	barrel	NNS
07-kim-ai	26	13	down	down	IN
07-kim-ai	26	14	a	a	DT
07-kim-ai	26	15	track	track	NN
07-kim-ai	26	16	where	where	WRB
07-kim-ai	26	17	five	five	CD
07-kim-ai	26	18	unsuspecting	unsuspecting	JJ
07-kim-ai	26	19	people	people	NNS
07-kim-ai	26	20	are	be	VBP
07-kim-ai	26	21	standing	stand	VBG
07-kim-ai	26	22	.	.	.
07-kim-ai	27	1	You	-PRON-	PRP
07-kim-ai	27	2	happen	happen	VBP
07-kim-ai	27	3	to	to	TO
07-kim-ai	27	4	be	be	VB
07-kim-ai	27	5	standing	stand	VBG
07-kim-ai	27	6	next	next	RB
07-kim-ai	27	7	to	to	IN
07-kim-ai	27	8	a	a	DT
07-kim-ai	27	9	lever	lever	NN
07-kim-ai	27	10	that	that	WDT
07-kim-ai	27	11	switches	switch	VBZ
07-kim-ai	27	12	the	the	DT
07-kim-ai	27	13	trolley	trolley	NN
07-kim-ai	27	14	onto	onto	IN
07-kim-ai	27	15	a	a	DT
07-kim-ai	27	16	different	different	JJ
07-kim-ai	27	17	track	track	NN
07-kim-ai	27	18	,	,	,
07-kim-ai	27	19	where	where	WRB
07-kim-ai	27	20	there	there	EX
07-kim-ai	27	21	is	be	VBZ
07-kim-ai	27	22	only	only	RB
07-kim-ai	27	23	one	one	CD
07-kim-ai	27	24	person	person	NN
07-kim-ai	27	25	.	.	.
07-kim-ai	28	1	Those	those	DT
07-kim-ai	28	2	who	who	WP
07-kim-ai	28	3	are	be	VBP
07-kim-ai	28	4	on	on	IN
07-kim-ai	28	5	either	either	DT
07-kim-ai	28	6	track	track	NN
07-kim-ai	28	7	will	will	MD
07-kim-ai	28	8	be	be	VB
07-kim-ai	28	9	killed	kill	VBN
07-kim-ai	28	10	if	if	IN
07-kim-ai	28	11	the	the	DT
07-kim-ai	28	12	trolley	trolley	NN
07-kim-ai	28	13	heads	head	VBZ
07-kim-ai	28	14	their	-PRON-	PRP$
07-kim-ai	28	15	way	way	NN
07-kim-ai	28	16	.	.	.
07-kim-ai	29	1	Should	Should	MD
07-kim-ai	29	2	you	-PRON-	PRP
07-kim-ai	29	3	pull	pull	VB
07-kim-ai	29	4	the	the	DT
07-kim-ai	29	5	lever	lever	NN
07-kim-ai	29	6	,	,	,
07-kim-ai	29	7	so	so	IN
07-kim-ai	29	8	that	that	IN
07-kim-ai	29	9	the	the	DT
07-kim-ai	29	10	runaway	runaway	JJ
07-kim-ai	29	11	trolley	trolley	NN
07-kim-ai	29	12	would	would	MD
07-kim-ai	29	13	kill	kill	VB
07-kim-ai	29	14	one	one	CD
07-kim-ai	29	15	per-	per-	NN
07-kim-ai	29	16	son	son	NN
07-kim-ai	29	17	instead	instead	RB
07-kim-ai	29	18	of	of	IN
07-kim-ai	29	19	five	five	CD
07-kim-ai	29	20	?	?	.
07-kim-ai	30	1	Unlike	unlike	IN
07-kim-ai	30	2	a	a	DT
07-kim-ai	30	3	person	person	NN
07-kim-ai	30	4	,	,	,
07-kim-ai	30	5	a	a	DT
07-kim-ai	30	6	machine	machine	NN
07-kim-ai	30	7	does	do	VBZ
07-kim-ai	30	8	not	not	RB
07-kim-ai	30	9	panic	panic	VB
07-kim-ai	30	10	or	or	CC
07-kim-ai	30	11	freeze	freeze	VB
07-kim-ai	30	12	and	and	CC
07-kim-ai	30	13	simply	simply	RB
07-kim-ai	30	14	follows	follow	VBZ
07-kim-ai	30	15	and	and	CC
07-kim-ai	30	16	executes	execute	VBZ
07-kim-ai	30	17	the	the	DT
07-kim-ai	30	18	given	give	VBN
07-kim-ai	30	19	instruction	instruction	NN
07-kim-ai	30	20	.	.	.
07-kim-ai	31	1	This	this	DT
07-kim-ai	31	2	means	mean	VBZ
07-kim-ai	31	3	that	that	IN
07-kim-ai	31	4	an	an	DT
07-kim-ai	31	5	AI	ai	NN
07-kim-ai	31	6	-	-	HYPH
07-kim-ai	31	7	powered	power	VBN
07-kim-ai	31	8	trolley	trolley	NN
07-kim-ai	31	9	may	may	MD
07-kim-ai	31	10	act	act	VB
07-kim-ai	31	11	morally	morally	RB
07-kim-ai	31	12	as	as	RB
07-kim-ai	31	13	long	long	RB
07-kim-ai	31	14	as	as	IN
07-kim-ai	31	15	it	-PRON-	PRP
07-kim-ai	31	16	is	be	VBZ
07-kim-ai	31	17	programmed	program	VBN
07-kim-ai	31	18	properly.6	properly.6	CD
07-kim-ai	31	19	The	the	DT
07-kim-ai	31	20	question	question	NN
07-kim-ai	31	21	itself	-PRON-	PRP
07-kim-ai	31	22	remains	remain	VBZ
07-kim-ai	31	23	,	,	,
07-kim-ai	31	24	however	however	RB
07-kim-ai	31	25	.	.	.
07-kim-ai	32	1	Should	Should	MD
07-kim-ai	32	2	the	the	DT
07-kim-ai	32	3	AI	ai	NN
07-kim-ai	32	4	-	-	HYPH
07-kim-ai	32	5	powered	power	VBN
07-kim-ai	32	6	trolley	trolley	NN
07-kim-ai	32	7	be	be	VB
07-kim-ai	32	8	programmed	program	VBN
07-kim-ai	32	9	to	to	TO
07-kim-ai	32	10	swerve	swerve	VB
07-kim-ai	32	11	or	or	CC
07-kim-ai	32	12	stay	stay	VB
07-kim-ai	32	13	on	on	RP
07-kim-ai	32	14	course	course	NN
07-kim-ai	32	15	?	?	.
07-kim-ai	33	1	Different	different	JJ
07-kim-ai	33	2	moral	moral	JJ
07-kim-ai	33	3	theories	theory	NNS
07-kim-ai	33	4	,	,	,
07-kim-ai	33	5	such	such	JJ
07-kim-ai	33	6	as	as	IN
07-kim-ai	33	7	virtue	virtue	NN
07-kim-ai	33	8	ethics	ethic	NNS
07-kim-ai	33	9	,	,	,
07-kim-ai	33	10	contractarianism	contractarianism	NN
07-kim-ai	33	11	,	,	,
07-kim-ai	33	12	and	and	CC
07-kim-ai	33	13	moral	moral	JJ
07-kim-ai	33	14	relativism	relativism	NN
07-kim-ai	33	15	,	,	,
07-kim-ai	33	16	take	take	VBP
07-kim-ai	33	17	different	different	JJ
07-kim-ai	33	18	positions	position	NNS
07-kim-ai	33	19	.	.	.
07-kim-ai	34	1	Here	here	RB
07-kim-ai	34	2	,	,	,
07-kim-ai	34	3	I	-PRON-	PRP
07-kim-ai	34	4	will	will	MD
07-kim-ai	34	5	consider	consider	VB
07-kim-ai	34	6	utilitarianism	utilitarianism	NN
07-kim-ai	34	7	and	and	CC
07-kim-ai	34	8	deontology	deontology	NN
07-kim-ai	34	9	.	.	.
07-kim-ai	35	1	Since	since	IN
07-kim-ai	35	2	their	-PRON-	PRP$
07-kim-ai	35	3	tenets	tenet	NNS
07-kim-ai	35	4	are	be	VBP
07-kim-ai	35	5	relatively	relatively	RB
07-kim-ai	35	6	straightforward	straightforward	JJ
07-kim-ai	35	7	,	,	,
07-kim-ai	35	8	most	most	JJS
07-kim-ai	35	9	AI	ai	NN
07-kim-ai	35	10	developers	developer	NNS
07-kim-ai	35	11	are	be	VBP
07-kim-ai	35	12	likely	likely	JJ
07-kim-ai	35	13	to	to	TO
07-kim-ai	35	14	look	look	VB
07-kim-ai	35	15	towards	towards	IN
07-kim-ai	35	16	those	those	DT
07-kim-ai	35	17	two	two	CD
07-kim-ai	35	18	moral	moral	JJ
07-kim-ai	35	19	theories	theory	NNS
07-kim-ai	35	20	for	for	IN
07-kim-ai	35	21	guidance	guidance	NN
07-kim-ai	35	22	and	and	CC
07-kim-ai	35	23	insight	insight	NN
07-kim-ai	35	24	.	.	.
07-kim-ai	36	1	Utilitarianism	Utilitarianism	NNP
07-kim-ai	36	2	argues	argue	VBZ
07-kim-ai	36	3	that	that	IN
07-kim-ai	36	4	the	the	DT
07-kim-ai	36	5	utility	utility	NN
07-kim-ai	36	6	of	of	IN
07-kim-ai	36	7	an	an	DT
07-kim-ai	36	8	action	action	NN
07-kim-ai	36	9	is	be	VBZ
07-kim-ai	36	10	what	what	WP
07-kim-ai	36	11	makes	make	VBZ
07-kim-ai	36	12	an	an	DT
07-kim-ai	36	13	action	action	NN
07-kim-ai	36	14	moral	moral	JJ
07-kim-ai	36	15	.	.	.
07-kim-ai	37	1	In	in	IN
07-kim-ai	37	2	this	this	DT
07-kim-ai	37	3	view	view	NN
07-kim-ai	37	4	,	,	,
07-kim-ai	37	5	what	what	WP
07-kim-ai	37	6	generates	generate	VBZ
07-kim-ai	37	7	the	the	DT
07-kim-ai	37	8	greatest	great	JJS
07-kim-ai	37	9	amount	amount	NN
07-kim-ai	37	10	of	of	IN
07-kim-ai	37	11	good	good	JJ
07-kim-ai	37	12	is	be	VBZ
07-kim-ai	37	13	the	the	DT
07-kim-ai	37	14	most	most	RBS
07-kim-ai	37	15	moral	moral	JJ
07-kim-ai	37	16	thing	thing	NN
07-kim-ai	37	17	to	to	TO
07-kim-ai	37	18	do	do	VB
07-kim-ai	37	19	.	.	.
07-kim-ai	38	1	If	if	IN
07-kim-ai	38	2	one	one	PRP
07-kim-ai	38	3	regards	regard	VBZ
07-kim-ai	38	4	five	five	CD
07-kim-ai	38	5	human	human	JJ
07-kim-ai	38	6	lives	life	NNS
07-kim-ai	38	7	as	as	IN
07-kim-ai	38	8	a	a	DT
07-kim-ai	38	9	greater	great	JJR
07-kim-ai	38	10	good	good	NN
07-kim-ai	38	11	than	than	IN
07-kim-ai	38	12	one	one	CD
07-kim-ai	38	13	,	,	,
07-kim-ai	38	14	then	then	RB
07-kim-ai	38	15	one	one	CD
07-kim-ai	38	16	acts	act	VBZ
07-kim-ai	38	17	morally	morally	RB
07-kim-ai	38	18	by	by	IN
07-kim-ai	38	19	pulling	pull	VBG
07-kim-ai	38	20	the	the	DT
07-kim-ai	38	21	lever	lever	NN
07-kim-ai	38	22	and	and	CC
07-kim-ai	38	23	diverting	divert	VBG
07-kim-ai	38	24	the	the	DT
07-kim-ai	38	25	trolley	trolley	NN
07-kim-ai	38	26	to	to	IN
07-kim-ai	38	27	the	the	DT
07-kim-ai	38	28	other	other	JJ
07-kim-ai	38	29	track	track	NN
07-kim-ai	38	30	.	.	.
07-kim-ai	39	1	By	by	IN
07-kim-ai	39	2	contrast	contrast	NN
07-kim-ai	39	3	,	,	,
07-kim-ai	39	4	deontology	deontology	NN
07-kim-ai	39	5	claims	claim	VBZ
07-kim-ai	39	6	that	that	IN
07-kim-ai	39	7	what	what	WP
07-kim-ai	39	8	determines	determine	VBZ
07-kim-ai	39	9	whether	whether	IN
07-kim-ai	39	10	an	an	DT
07-kim-ai	39	11	action	action	NN
07-kim-ai	39	12	is	be	VBZ
07-kim-ai	39	13	morally	morally	RB
07-kim-ai	39	14	right	right	JJ
07-kim-ai	39	15	or	or	CC
07-kim-ai	39	16	wrong	wrong	JJ
07-kim-ai	39	17	is	be	VBZ
07-kim-ai	39	18	not	not	RB
07-kim-ai	39	19	its	-PRON-	PRP$
07-kim-ai	39	20	utility	utility	NN
07-kim-ai	39	21	but	but	CC
07-kim-ai	39	22	moral	moral	JJ
07-kim-ai	39	23	rules	rule	NNS
07-kim-ai	39	24	.	.	.
07-kim-ai	40	1	If	if	IN
07-kim-ai	40	2	an	an	DT
07-kim-ai	40	3	action	action	NN
07-kim-ai	40	4	is	be	VBZ
07-kim-ai	40	5	in	in	IN
07-kim-ai	40	6	accordance	accordance	NN
07-kim-ai	40	7	with	with	IN
07-kim-ai	40	8	those	those	DT
07-kim-ai	40	9	rules	rule	NNS
07-kim-ai	40	10	,	,	,
07-kim-ai	40	11	then	then	RB
07-kim-ai	40	12	the	the	DT
07-kim-ai	40	13	action	action	NN
07-kim-ai	40	14	is	be	VBZ
07-kim-ai	40	15	morally	morally	RB
07-kim-ai	40	16	right	right	JJ
07-kim-ai	40	17	.	.	.
07-kim-ai	41	1	Otherwise	otherwise	RB
07-kim-ai	41	2	,	,	,
07-kim-ai	41	3	it	-PRON-	PRP
07-kim-ai	41	4	is	be	VBZ
07-kim-ai	41	5	morally	morally	RB
07-kim-ai	41	6	4Machine	4machine	CD
07-kim-ai	41	7	learning	learning	JJ
07-kim-ai	41	8	and	and	CC
07-kim-ai	41	9	deep	deep	JJ
07-kim-ai	41	10	learning	learning	NN
07-kim-ai	41	11	have	have	VBP
07-kim-ai	41	12	gained	gain	VBN
07-kim-ai	41	13	momentum	momentum	NN
07-kim-ai	41	14	because	because	IN
07-kim-ai	41	15	the	the	DT
07-kim-ai	41	16	cost	cost	NN
07-kim-ai	41	17	of	of	IN
07-kim-ai	41	18	high	high	JJ
07-kim-ai	41	19	-	-	HYPH
07-kim-ai	41	20	performance	performance	NN
07-kim-ai	41	21	computing	computing	NN
07-kim-ai	41	22	has	have	VBZ
07-kim-ai	41	23	significantly	significantly	RB
07-kim-ai	41	24	decreased	decrease	VBN
07-kim-ai	41	25	and	and	CC
07-kim-ai	41	26	large	large	JJ
07-kim-ai	41	27	data	data	NN
07-kim-ai	41	28	sets	set	NNS
07-kim-ai	41	29	have	have	VBP
07-kim-ai	41	30	become	become	VBN
07-kim-ai	41	31	more	more	RBR
07-kim-ai	41	32	widely	widely	RB
07-kim-ai	41	33	available	available	JJ
07-kim-ai	41	34	.	.	.
07-kim-ai	42	1	For	for	IN
07-kim-ai	42	2	example	example	NN
07-kim-ai	42	3	,	,	,
07-kim-ai	42	4	the	the	DT
07-kim-ai	42	5	data	datum	NNS
07-kim-ai	42	6	in	in	IN
07-kim-ai	42	7	the	the	DT
07-kim-ai	42	8	ImageNet	ImageNet	NNP
07-kim-ai	42	9	contains	contain	VBZ
07-kim-ai	42	10	more	more	JJR
07-kim-ai	42	11	than	than	IN
07-kim-ai	42	12	14	14	CD
07-kim-ai	42	13	million	million	CD
07-kim-ai	42	14	hand	hand	NN
07-kim-ai	42	15	-	-	HYPH
07-kim-ai	42	16	annotated	annotate	VBN
07-kim-ai	42	17	images	image	NNS
07-kim-ai	42	18	.	.	.
07-kim-ai	43	1	The	the	DT
07-kim-ai	43	2	ImageNet	ImageNet	NNP
07-kim-ai	43	3	data	datum	NNS
07-kim-ai	43	4	have	have	VBP
07-kim-ai	43	5	been	be	VBN
07-kim-ai	43	6	used	use	VBN
07-kim-ai	43	7	for	for	IN
07-kim-ai	43	8	the	the	DT
07-kim-ai	43	9	well	well	RB
07-kim-ai	43	10	-	-	HYPH
07-kim-ai	43	11	known	know	VBN
07-kim-ai	43	12	annual	annual	JJ
07-kim-ai	43	13	AI	AI	NNP
07-kim-ai	43	14	competition	competition	NN
07-kim-ai	43	15	for	for	IN
07-kim-ai	43	16	object	object	NN
07-kim-ai	43	17	detection	detection	NN
07-kim-ai	43	18	and	and	CC
07-kim-ai	43	19	image	image	NN
07-kim-ai	43	20	classification	classification	NN
07-kim-ai	43	21	at	at	IN
07-kim-ai	43	22	large	large	JJ
07-kim-ai	43	23	scale	scale	NN
07-kim-ai	43	24	from	from	IN
07-kim-ai	43	25	2010	2010	CD
07-kim-ai	43	26	to	to	IN
07-kim-ai	43	27	2017	2017	CD
07-kim-ai	43	28	.	.	.
07-kim-ai	44	1	See	see	VB
07-kim-ai	44	2	?	?	.
07-kim-ai	44	3	iiT	iiT	VBN
07-kim-ai	44	4	,	,	,
07-kim-ai	44	5	ffrrrXBK	ffrrrXBK	NNP
07-kim-ai	44	6	�	�	NNP
07-kim-ai	44	7	;	;	:
07-kim-ai	44	8	2@M2iXQ`;f+?	2@M2iXQ`;f+?	NNP
07-kim-ai	44	9	�	�	NNP
07-kim-ai	44	10	HH2M;2bfGao_*f	HH2M;2bfGao_*f	NNP
07-kim-ai	44	11	.	.	.
07-kim-ai	44	12	5For	5for	IN
07-kim-ai	44	13	an	an	DT
07-kim-ai	44	14	excellent	excellent	JJ
07-kim-ai	44	15	history	history	NN
07-kim-ai	44	16	of	of	IN
07-kim-ai	44	17	AI	AI	NNP
07-kim-ai	44	18	research	research	NN
07-kim-ai	44	19	,	,	,
07-kim-ai	44	20	see	see	VB
07-kim-ai	44	21	chapter	chapter	NN
07-kim-ai	44	22	1	1	CD
07-kim-ai	44	23	,	,	,
07-kim-ai	44	24	“	"	``
07-kim-ai	44	25	What	what	WP
07-kim-ai	44	26	is	be	VBZ
07-kim-ai	44	27	Artificial	Artificial	NNP
07-kim-ai	44	28	Intelligence	Intelligence	NNP
07-kim-ai	44	29	,	,	,
07-kim-ai	44	30	”	"	''
07-kim-ai	44	31	of	of	IN
07-kim-ai	44	32	Boden	Boden	NNP
07-kim-ai	44	33	2016	2016	CD
07-kim-ai	44	34	,	,	,
07-kim-ai	44	35	1	1	CD
07-kim-ai	44	36	-	-	SYM
07-kim-ai	44	37	20	20	CD
07-kim-ai	44	38	.	.	.
07-kim-ai	45	1	6Programming	6programming	NN
07-kim-ai	45	2	here	here	RB
07-kim-ai	45	3	does	do	VBZ
07-kim-ai	45	4	not	not	RB
07-kim-ai	45	5	exclusively	exclusively	RB
07-kim-ai	45	6	refer	refer	VB
07-kim-ai	45	7	to	to	IN
07-kim-ai	45	8	a	a	DT
07-kim-ai	45	9	deep	deep	JJ
07-kim-ai	45	10	learning	learning	NN
07-kim-ai	45	11	or	or	CC
07-kim-ai	45	12	machine	machine	NN
07-kim-ai	45	13	learning	learning	NN
07-kim-ai	45	14	approach	approach	NN
07-kim-ai	45	15	.	.	.
07-kim-ai	46	1	http://www.image-net.org/challenges/LSVRC/	http://www.image-net.org/challenges/LSVRC/	NNP
07-kim-ai	46	2	http://www.image-net.org/challenges/LSVRC/	http://www.image-net.org/challenges/LSVRC/	NNP
07-kim-ai	46	3	Kim	Kim	NNP
07-kim-ai	46	4	75	75	CD
07-kim-ai	46	5	wrong	wrong	NN
07-kim-ai	46	6	.	.	.
07-kim-ai	47	1	If	if	IN
07-kim-ai	47	2	not	not	RB
07-kim-ai	47	3	to	to	TO
07-kim-ai	47	4	kill	kill	VB
07-kim-ai	47	5	another	another	DT
07-kim-ai	47	6	human	human	NN
07-kim-ai	47	7	being	being	NN
07-kim-ai	47	8	is	be	VBZ
07-kim-ai	47	9	one	one	CD
07-kim-ai	47	10	of	of	IN
07-kim-ai	47	11	those	those	DT
07-kim-ai	47	12	moral	moral	JJ
07-kim-ai	47	13	rules	rule	NNS
07-kim-ai	47	14	,	,	,
07-kim-ai	47	15	then	then	RB
07-kim-ai	47	16	killing	kill	VBG
07-kim-ai	47	17	someone	someone	NN
07-kim-ai	47	18	is	be	VBZ
07-kim-ai	47	19	morally	morally	RB
07-kim-ai	47	20	wrong	wrong	JJ
07-kim-ai	47	21	even	even	RB
07-kim-ai	47	22	if	if	IN
07-kim-ai	47	23	it	-PRON-	PRP
07-kim-ai	47	24	is	be	VBZ
07-kim-ai	47	25	to	to	TO
07-kim-ai	47	26	save	save	VB
07-kim-ai	47	27	more	more	JJR
07-kim-ai	47	28	lives	life	NNS
07-kim-ai	47	29	.	.	.
07-kim-ai	48	1	Note	note	VB
07-kim-ai	48	2	that	that	IN
07-kim-ai	48	3	these	these	DT
07-kim-ai	48	4	are	be	VBP
07-kim-ai	48	5	highly	highly	RB
07-kim-ai	48	6	simplified	simplified	JJ
07-kim-ai	48	7	accounts	account	NNS
07-kim-ai	48	8	of	of	IN
07-kim-ai	48	9	utilitarianism	utilitarianism	NN
07-kim-ai	48	10	and	and	CC
07-kim-ai	48	11	deontology	deontology	NN
07-kim-ai	48	12	.	.	.
07-kim-ai	49	1	The	the	DT
07-kim-ai	49	2	good	good	NN
07-kim-ai	49	3	in	in	IN
07-kim-ai	49	4	utilitarianism	utilitarianism	NN
07-kim-ai	49	5	can	can	MD
07-kim-ai	49	6	be	be	VB
07-kim-ai	49	7	interpreted	interpret	VBN
07-kim-ai	49	8	in	in	IN
07-kim-ai	49	9	many	many	JJ
07-kim-ai	49	10	different	different	JJ
07-kim-ai	49	11	ways	way	NNS
07-kim-ai	49	12	,	,	,
07-kim-ai	49	13	and	and	CC
07-kim-ai	49	14	the	the	DT
07-kim-ai	49	15	issue	issue	NN
07-kim-ai	49	16	of	of	IN
07-kim-ai	49	17	conflicting	conflict	VBG
07-kim-ai	49	18	moral	moral	JJ
07-kim-ai	49	19	rules	rule	NNS
07-kim-ai	49	20	is	be	VBZ
07-kim-ai	49	21	a	a	DT
07-kim-ai	49	22	perennial	perennial	JJ
07-kim-ai	49	23	problem	problem	NN
07-kim-ai	49	24	that	that	IN
07-kim-ai	49	25	deontological	deontological	JJ
07-kim-ai	49	26	ethics	ethic	NNS
07-kim-ai	49	27	grapples	grapple	VBZ
07-kim-ai	49	28	with.7	with.7	CD
07-kim-ai	49	29	For	for	IN
07-kim-ai	49	30	our	-PRON-	PRP$
07-kim-ai	49	31	purpose	purpose	NN
07-kim-ai	49	32	,	,	,
07-kim-ai	49	33	however	however	RB
07-kim-ai	49	34	,	,	,
07-kim-ai	49	35	these	these	DT
07-kim-ai	49	36	simplified	simplify	VBN
07-kim-ai	49	37	accounts	account	NNS
07-kim-ai	49	38	are	be	VBP
07-kim-ai	49	39	sufficient	sufficient	JJ
07-kim-ai	49	40	to	to	TO
07-kim-ai	49	41	highlight	highlight	VB
07-kim-ai	49	42	the	the	DT
07-kim-ai	49	43	aspects	aspect	NNS
07-kim-ai	49	44	in	in	IN
07-kim-ai	49	45	which	which	WDT
07-kim-ai	49	46	the	the	DT
07-kim-ai	49	47	utilitarian	utilitarian	NN
07-kim-ai	49	48	and	and	CC
07-kim-ai	49	49	the	the	DT
07-kim-ai	49	50	deontological	deontological	JJ
07-kim-ai	49	51	position	position	NN
07-kim-ai	49	52	appeal	appeal	VBP
07-kim-ai	49	53	to	to	IN
07-kim-ai	49	54	and	and	CC
07-kim-ai	49	55	go	go	VB
07-kim-ai	49	56	against	against	IN
07-kim-ai	49	57	our	-PRON-	PRP$
07-kim-ai	49	58	moral	moral	JJ
07-kim-ai	49	59	intuition	intuition	NN
07-kim-ai	49	60	at	at	IN
07-kim-ai	49	61	the	the	DT
07-kim-ai	49	62	same	same	JJ
07-kim-ai	49	63	time	time	NN
07-kim-ai	49	64	.	.	.
07-kim-ai	50	1	If	if	IN
07-kim-ai	50	2	a	a	DT
07-kim-ai	50	3	trolley	trolley	NN
07-kim-ai	50	4	can	can	MD
07-kim-ai	50	5	not	not	RB
07-kim-ai	50	6	be	be	VB
07-kim-ai	50	7	stopped	stop	VBN
07-kim-ai	50	8	,	,	,
07-kim-ai	50	9	saving	save	VBG
07-kim-ai	50	10	five	five	CD
07-kim-ai	50	11	lives	life	NNS
07-kim-ai	50	12	over	over	IN
07-kim-ai	50	13	one	one	NN
07-kim-ai	50	14	seems	seem	VBZ
07-kim-ai	50	15	to	to	TO
07-kim-ai	50	16	be	be	VB
07-kim-ai	50	17	a	a	DT
07-kim-ai	50	18	right	right	JJ
07-kim-ai	50	19	thing	thing	NN
07-kim-ai	50	20	to	to	TO
07-kim-ai	50	21	do	do	VB
07-kim-ai	50	22	.	.	.
07-kim-ai	51	1	Util-	util-	JJ
07-kim-ai	51	2	itarianism	itarianism	NN
07-kim-ai	51	3	appears	appear	VBZ
07-kim-ai	51	4	to	to	TO
07-kim-ai	51	5	get	get	VB
07-kim-ai	51	6	things	thing	NNS
07-kim-ai	51	7	right	right	JJ
07-kim-ai	51	8	in	in	IN
07-kim-ai	51	9	this	this	DT
07-kim-ai	51	10	respect	respect	NN
07-kim-ai	51	11	.	.	.
07-kim-ai	52	1	However	however	RB
07-kim-ai	52	2	,	,	,
07-kim-ai	52	3	it	-PRON-	PRP
07-kim-ai	52	4	is	be	VBZ
07-kim-ai	52	5	hard	hard	JJ
07-kim-ai	52	6	to	to	TO
07-kim-ai	52	7	dispute	dispute	VB
07-kim-ai	52	8	that	that	IN
07-kim-ai	52	9	killing	kill	VBG
07-kim-ai	52	10	people	people	NNS
07-kim-ai	52	11	is	be	VBZ
07-kim-ai	52	12	wrong	wrong	JJ
07-kim-ai	52	13	.	.	.
07-kim-ai	53	1	If	if	IN
07-kim-ai	53	2	killing	killing	NN
07-kim-ai	53	3	is	be	VBZ
07-kim-ai	53	4	morally	morally	RB
07-kim-ai	53	5	wrong	wrong	JJ
07-kim-ai	53	6	no	no	RB
07-kim-ai	53	7	matter	matter	RB
07-kim-ai	53	8	what	what	WP
07-kim-ai	53	9	,	,	,
07-kim-ai	53	10	deontology	deontology	NN
07-kim-ai	53	11	seems	seem	VBZ
07-kim-ai	53	12	to	to	TO
07-kim-ai	53	13	make	make	VB
07-kim-ai	53	14	more	more	JJR
07-kim-ai	53	15	sense	sense	NN
07-kim-ai	53	16	.	.	.
07-kim-ai	54	1	With	with	IN
07-kim-ai	54	2	moral	moral	JJ
07-kim-ai	54	3	theories	theory	NNS
07-kim-ai	54	4	,	,	,
07-kim-ai	54	5	things	thing	NNS
07-kim-ai	54	6	seem	seem	VBP
07-kim-ai	54	7	to	to	TO
07-kim-ai	54	8	get	get	VB
07-kim-ai	54	9	more	more	RBR
07-kim-ai	54	10	confusing	confusing	JJ
07-kim-ai	54	11	.	.	.
07-kim-ai	55	1	Furthermore	furthermore	RB
07-kim-ai	55	2	,	,	,
07-kim-ai	55	3	consider	consider	VB
07-kim-ai	55	4	the	the	DT
07-kim-ai	55	5	case	case	NN
07-kim-ai	55	6	in	in	IN
07-kim-ai	55	7	which	which	WDT
07-kim-ai	55	8	one	one	NN
07-kim-ai	55	9	freezes	freeze	VBZ
07-kim-ai	55	10	and	and	CC
07-kim-ai	55	11	fails	fail	VBZ
07-kim-ai	55	12	to	to	TO
07-kim-ai	55	13	pull	pull	VB
07-kim-ai	55	14	the	the	DT
07-kim-ai	55	15	lever	lever	NN
07-kim-ai	55	16	.	.	.
07-kim-ai	56	1	According	accord	VBG
07-kim-ai	56	2	to	to	IN
07-kim-ai	56	3	utilitarianism	utilitarianism	NN
07-kim-ai	56	4	,	,	,
07-kim-ai	56	5	this	this	DT
07-kim-ai	56	6	would	would	MD
07-kim-ai	56	7	be	be	VB
07-kim-ai	56	8	morally	morally	RB
07-kim-ai	56	9	wrong	wrong	JJ
07-kim-ai	56	10	because	because	IN
07-kim-ai	56	11	it	-PRON-	PRP
07-kim-ai	56	12	fails	fail	VBZ
07-kim-ai	56	13	to	to	TO
07-kim-ai	56	14	maximize	maximize	VB
07-kim-ai	56	15	the	the	DT
07-kim-ai	56	16	greatest	great	JJS
07-kim-ai	56	17	good	good	JJ
07-kim-ai	56	18	,	,	,
07-kim-ai	56	19	i.e.	i.e.	FW
07-kim-ai	57	1	human	human	JJ
07-kim-ai	57	2	lives	life	NNS
07-kim-ai	57	3	.	.	.
07-kim-ai	58	1	But	but	CC
07-kim-ai	58	2	how	how	WRB
07-kim-ai	58	3	far	far	RB
07-kim-ai	58	4	should	should	MD
07-kim-ai	58	5	one	one	PRP
07-kim-ai	58	6	go	go	VB
07-kim-ai	58	7	to	to	TO
07-kim-ai	58	8	maximize	maximize	VB
07-kim-ai	58	9	the	the	DT
07-kim-ai	58	10	good	good	NN
07-kim-ai	58	11	?	?	.
07-kim-ai	59	1	Suppose	suppose	VB
07-kim-ai	59	2	there	there	EX
07-kim-ai	59	3	is	be	VBZ
07-kim-ai	59	4	a	a	DT
07-kim-ai	59	5	very	very	RB
07-kim-ai	59	6	large	large	JJ
07-kim-ai	59	7	person	person	NN
07-kim-ai	59	8	on	on	IN
07-kim-ai	59	9	a	a	DT
07-kim-ai	59	10	footbridge	footbridge	NN
07-kim-ai	59	11	over	over	IN
07-kim-ai	59	12	the	the	DT
07-kim-ai	59	13	trolley	trolley	NN
07-kim-ai	59	14	track	track	NN
07-kim-ai	59	15	,	,	,
07-kim-ai	59	16	and	and	CC
07-kim-ai	59	17	one	one	NN
07-kim-ai	59	18	pushes	push	VBZ
07-kim-ai	59	19	that	that	DT
07-kim-ai	59	20	person	person	NN
07-kim-ai	59	21	off	off	IN
07-kim-ai	59	22	the	the	DT
07-kim-ai	59	23	footbridge	footbridge	NN
07-kim-ai	59	24	onto	onto	IN
07-kim-ai	59	25	the	the	DT
07-kim-ai	59	26	track	track	NN
07-kim-ai	59	27	,	,	,
07-kim-ai	59	28	thus	thus	RB
07-kim-ai	59	29	stopping	stop	VBG
07-kim-ai	59	30	the	the	DT
07-kim-ai	59	31	trolley	trolley	NN
07-kim-ai	59	32	and	and	CC
07-kim-ai	59	33	saving	save	VBG
07-kim-ai	59	34	the	the	DT
07-kim-ai	59	35	five	five	CD
07-kim-ai	59	36	people	people	NNS
07-kim-ai	59	37	.	.	.
07-kim-ai	60	1	Would	Would	MD
07-kim-ai	60	2	this	this	DT
07-kim-ai	60	3	count	count	VB
07-kim-ai	60	4	as	as	IN
07-kim-ai	60	5	a	a	DT
07-kim-ai	60	6	right	right	JJ
07-kim-ai	60	7	thing	thing	NN
07-kim-ai	60	8	to	to	TO
07-kim-ai	60	9	do	do	VB
07-kim-ai	60	10	?	?	.
07-kim-ai	61	1	Utilitarianism	Utilitarianism	NNP
07-kim-ai	61	2	may	may	MD
07-kim-ai	61	3	argue	argue	VB
07-kim-ai	61	4	that	that	DT
07-kim-ai	61	5	.	.	.
07-kim-ai	62	1	But	but	CC
07-kim-ai	62	2	in	in	IN
07-kim-ai	62	3	real	real	JJ
07-kim-ai	62	4	life	life	NN
07-kim-ai	62	5	,	,	,
07-kim-ai	62	6	many	many	JJ
07-kim-ai	62	7	would	would	MD
07-kim-ai	62	8	consider	consider	VB
07-kim-ai	62	9	throwing	throw	VBG
07-kim-ai	62	10	a	a	DT
07-kim-ai	62	11	person	person	NN
07-kim-ai	62	12	morally	morally	RB
07-kim-ai	62	13	wrong	wrong	JJ
07-kim-ai	62	14	but	but	CC
07-kim-ai	62	15	pulling	pull	VBG
07-kim-ai	62	16	the	the	DT
07-kim-ai	62	17	lever	lever	NN
07-kim-ai	62	18	morally	morally	RB
07-kim-ai	62	19	permissible.8	permissible.8	VBP
07-kim-ai	62	20	The	the	DT
07-kim-ai	62	21	problem	problem	NN
07-kim-ai	62	22	with	with	IN
07-kim-ai	62	23	utilitarianism	utilitarianism	NN
07-kim-ai	62	24	is	be	VBZ
07-kim-ai	62	25	that	that	IN
07-kim-ai	62	26	it	-PRON-	PRP
07-kim-ai	62	27	treats	treat	VBZ
07-kim-ai	62	28	the	the	DT
07-kim-ai	62	29	good	good	JJ
07-kim-ai	62	30	as	as	IN
07-kim-ai	62	31	something	something	NN
07-kim-ai	62	32	inherently	inherently	RB
07-kim-ai	62	33	quantifi-	quantifi-	VBD
07-kim-ai	62	34	able	able	JJ
07-kim-ai	62	35	,	,	,
07-kim-ai	62	36	comparable	comparable	JJ
07-kim-ai	62	37	,	,	,
07-kim-ai	62	38	calculable	calculable	JJ
07-kim-ai	62	39	,	,	,
07-kim-ai	62	40	and	and	CC
07-kim-ai	62	41	additive	additive	NNP
07-kim-ai	62	42	.	.	.
07-kim-ai	63	1	But	but	CC
07-kim-ai	63	2	not	not	RB
07-kim-ai	63	3	all	all	DT
07-kim-ai	63	4	considerations	consideration	NNS
07-kim-ai	63	5	that	that	WDT
07-kim-ai	63	6	we	-PRON-	PRP
07-kim-ai	63	7	have	have	VBP
07-kim-ai	63	8	to	to	TO
07-kim-ai	63	9	factor	factor	VB
07-kim-ai	63	10	into	into	IN
07-kim-ai	63	11	moral	moral	JJ
07-kim-ai	63	12	decision	decision	NN
07-kim-ai	63	13	-	-	HYPH
07-kim-ai	63	14	making	making	NN
07-kim-ai	63	15	are	be	VBP
07-kim-ai	63	16	measurable	measurable	JJ
07-kim-ai	63	17	in	in	IN
07-kim-ai	63	18	numbers	number	NNS
07-kim-ai	63	19	.	.	.
07-kim-ai	64	1	What	what	WP
07-kim-ai	64	2	if	if	IN
07-kim-ai	64	3	the	the	DT
07-kim-ai	64	4	five	five	CD
07-kim-ai	64	5	people	people	NNS
07-kim-ai	64	6	on	on	IN
07-kim-ai	64	7	the	the	DT
07-kim-ai	64	8	track	track	NN
07-kim-ai	64	9	are	be	VBP
07-kim-ai	64	10	help-	help-	VBN
07-kim-ai	64	11	less	less	JJR
07-kim-ai	64	12	babies	baby	NNS
07-kim-ai	64	13	or	or	CC
07-kim-ai	64	14	murderers	murderer	NNS
07-kim-ai	64	15	who	who	WP
07-kim-ai	64	16	just	just	RB
07-kim-ai	64	17	escaped	escape	VBD
07-kim-ai	64	18	from	from	IN
07-kim-ai	64	19	the	the	DT
07-kim-ai	64	20	prison	prison	NN
07-kim-ai	64	21	?	?	.
07-kim-ai	65	1	Would	Would	MD
07-kim-ai	65	2	or	or	CC
07-kim-ai	65	3	should	should	MD
07-kim-ai	65	4	that	that	DT
07-kim-ai	65	5	affect	affect	VB
07-kim-ai	65	6	our	-PRON-	PRP$
07-kim-ai	65	7	de-	de-	NN
07-kim-ai	65	8	cision	cision	NN
07-kim-ai	65	9	?	?	.
07-kim-ai	66	1	Some	some	DT
07-kim-ai	66	2	of	of	IN
07-kim-ai	66	3	us	-PRON-	PRP
07-kim-ai	66	4	would	would	MD
07-kim-ai	66	5	surely	surely	RB
07-kim-ai	66	6	hesitate	hesitate	VB
07-kim-ai	66	7	to	to	TO
07-kim-ai	66	8	save	save	VB
07-kim-ai	66	9	the	the	DT
07-kim-ai	66	10	lives	life	NNS
07-kim-ai	66	11	of	of	IN
07-kim-ai	66	12	five	five	CD
07-kim-ai	66	13	murderers	murderer	NNS
07-kim-ai	66	14	by	by	IN
07-kim-ai	66	15	sacrificing	sacrifice	VBG
07-kim-ai	66	16	one	one	CD
07-kim-ai	66	17	innocent	innocent	JJ
07-kim-ai	66	18	baby	baby	NN
07-kim-ai	66	19	.	.	.
07-kim-ai	67	1	But	but	CC
07-kim-ai	67	2	what	what	WP
07-kim-ai	67	3	if	if	IN
07-kim-ai	67	4	things	thing	NNS
07-kim-ai	67	5	were	be	VBD
07-kim-ai	67	6	different	different	JJ
07-kim-ai	67	7	and	and	CC
07-kim-ai	67	8	we	-PRON-	PRP
07-kim-ai	67	9	were	be	VBD
07-kim-ai	67	10	comparing	compare	VBG
07-kim-ai	67	11	five	five	CD
07-kim-ai	67	12	school	school	NN
07-kim-ai	67	13	children	child	NNS
07-kim-ai	67	14	ver-	ver-	RB
07-kim-ai	67	15	sus	sus	NNP
07-kim-ai	67	16	one	one	CD
07-kim-ai	67	17	baby	baby	NN
07-kim-ai	67	18	or	or	CC
07-kim-ai	67	19	five	five	CD
07-kim-ai	67	20	babies	baby	NNS
07-kim-ai	67	21	versus	versus	IN
07-kim-ai	67	22	one	one	CD
07-kim-ai	67	23	school	school	NN
07-kim-ai	67	24	child	child	NN
07-kim-ai	67	25	?	?	.
07-kim-ai	68	1	No	no	DT
07-kim-ai	68	2	one	one	NN
07-kim-ai	68	3	can	can	MD
07-kim-ai	68	4	say	say	VB
07-kim-ai	68	5	for	for	IN
07-kim-ai	68	6	sure	sure	JJ
07-kim-ai	68	7	what	what	WP
07-kim-ai	68	8	is	be	VBZ
07-kim-ai	68	9	the	the	DT
07-kim-ai	68	10	morally	morally	RB
07-kim-ai	68	11	right	right	JJ
07-kim-ai	68	12	action	action	NN
07-kim-ai	68	13	in	in	IN
07-kim-ai	68	14	those	those	DT
07-kim-ai	68	15	cases.9	cases.9	NNS
07-kim-ai	68	16	While	while	IN
07-kim-ai	68	17	the	the	DT
07-kim-ai	68	18	utilitarian	utilitarian	JJ
07-kim-ai	68	19	position	position	NN
07-kim-ai	68	20	appears	appear	VBZ
07-kim-ai	68	21	less	less	RBR
07-kim-ai	68	22	persuasive	persuasive	JJ
07-kim-ai	68	23	in	in	IN
07-kim-ai	68	24	light	light	NN
07-kim-ai	68	25	of	of	IN
07-kim-ai	68	26	these	these	DT
07-kim-ai	68	27	considerations	consideration	NNS
07-kim-ai	68	28	,	,	,
07-kim-ai	68	29	deon-	deon-	NN
07-kim-ai	68	30	tology	tology	NN
07-kim-ai	68	31	does	do	VBZ
07-kim-ai	68	32	n’t	not	RB
07-kim-ai	68	33	fare	fare	VB
07-kim-ai	68	34	too	too	RB
07-kim-ai	68	35	well	well	RB
07-kim-ai	68	36	,	,	,
07-kim-ai	68	37	either	either	RB
07-kim-ai	68	38	.	.	.
07-kim-ai	69	1	Deontology	deontology	NN
07-kim-ai	69	2	emphasizes	emphasize	VBZ
07-kim-ai	69	3	one	one	PRP
07-kim-ai	69	4	’s	’s	NN
07-kim-ai	69	5	duty	duty	NN
07-kim-ai	69	6	to	to	TO
07-kim-ai	69	7	observe	observe	VB
07-kim-ai	69	8	moral	moral	JJ
07-kim-ai	69	9	rules	rule	NNS
07-kim-ai	69	10	.	.	.
07-kim-ai	70	1	But	but	CC
07-kim-ai	70	2	what	what	WP
07-kim-ai	70	3	if	if	IN
07-kim-ai	70	4	those	those	DT
07-kim-ai	70	5	moral	moral	JJ
07-kim-ai	70	6	rules	rule	NNS
07-kim-ai	70	7	conflict	conflict	NN
07-kim-ai	70	8	with	with	IN
07-kim-ai	70	9	one	one	CD
07-kim-ai	70	10	another	another	DT
07-kim-ai	70	11	?	?	.
07-kim-ai	71	1	Between	between	IN
07-kim-ai	71	2	the	the	DT
07-kim-ai	71	3	two	two	CD
07-kim-ai	71	4	moral	moral	JJ
07-kim-ai	71	5	rules	rule	NNS
07-kim-ai	71	6	,	,	,
07-kim-ai	71	7	“	"	``
07-kim-ai	71	8	do	do	VB
07-kim-ai	71	9	not	not	RB
07-kim-ai	71	10	kill	kill	VB
07-kim-ai	71	11	a	a	DT
07-kim-ai	71	12	person	person	NN
07-kim-ai	71	13	”	"	''
07-kim-ai	71	14	and	and	CC
07-kim-ai	71	15	“	"	``
07-kim-ai	71	16	save	save	VB
07-kim-ai	71	17	lives	life	NNS
07-kim-ai	71	18	,	,	,
07-kim-ai	71	19	”	"	''
07-kim-ai	71	20	which	which	WDT
07-kim-ai	71	21	one	one	PRP
07-kim-ai	71	22	should	should	MD
07-kim-ai	71	23	trump	trump	VB
07-kim-ai	71	24	the	the	DT
07-kim-ai	71	25	other	other	JJ
07-kim-ai	71	26	?	?	.
07-kim-ai	72	1	The	the	DT
07-kim-ai	72	2	conflict	conflict	NN
07-kim-ai	72	3	among	among	IN
07-kim-ai	72	4	values	value	NNS
07-kim-ai	72	5	is	be	VBZ
07-kim-ai	72	6	common	common	JJ
07-kim-ai	72	7	in	in	IN
07-kim-ai	72	8	life	life	NN
07-kim-ai	72	9	,	,	,
07-kim-ai	72	10	and	and	CC
07-kim-ai	72	11	deontology	deontology	NN
07-kim-ai	72	12	faces	face	VBZ
07-kim-ai	72	13	difficulty	difficulty	NN
07-kim-ai	72	14	in	in	IN
07-kim-ai	72	15	guiding	guide	VBG
07-kim-ai	72	16	how	how	WRB
07-kim-ai	72	17	an	an	DT
07-kim-ai	72	18	intelligent	intelligent	JJ
07-kim-ai	72	19	agent	agent	NN
07-kim-ai	72	20	is	be	VBZ
07-kim-ai	72	21	to	to	TO
07-kim-ai	72	22	act	act	VB
07-kim-ai	72	23	in	in	IN
07-kim-ai	72	24	a	a	DT
07-kim-ai	72	25	tricky	tricky	JJ
07-kim-ai	72	26	situation	situation	NN
07-kim-ai	72	27	such	such	JJ
07-kim-ai	72	28	as	as	IN
07-kim-ai	72	29	the	the	DT
07-kim-ai	72	30	trolley	trolley	NN
07-kim-ai	72	31	problem.10	problem.10	NNP
07-kim-ai	72	32	Understanding	Understanding	NNP
07-kim-ai	72	33	What	what	WP
07-kim-ai	72	34	Ethics	Ethics	NNP
07-kim-ai	72	35	Has	have	VBZ
07-kim-ai	72	36	to	to	TO
07-kim-ai	72	37	Offer	offer	VB
07-kim-ai	72	38	Now	now	RB
07-kim-ai	72	39	,	,	,
07-kim-ai	72	40	let	let	VB
07-kim-ai	72	41	us	-PRON-	PRP
07-kim-ai	72	42	consider	consider	VB
07-kim-ai	72	43	AI	AI	NNP
07-kim-ai	72	44	-	-	HYPH
07-kim-ai	72	45	powered	power	VBN
07-kim-ai	72	46	military	military	JJ
07-kim-ai	72	47	robots	robot	NNS
07-kim-ai	72	48	and	and	CC
07-kim-ai	72	49	autonomous	autonomous	JJ
07-kim-ai	72	50	weapons	weapon	NNS
07-kim-ai	72	51	systems	system	NNS
07-kim-ai	72	52	since	since	IN
07-kim-ai	72	53	they	-PRON-	PRP
07-kim-ai	72	54	present	present	VBP
07-kim-ai	72	55	the	the	DT
07-kim-ai	72	56	moral	moral	JJ
07-kim-ai	72	57	dilemma	dilemma	NN
07-kim-ai	72	58	in	in	IN
07-kim-ai	72	59	the	the	DT
07-kim-ai	72	60	trolley	trolley	NN
07-kim-ai	72	61	problem	problem	NN
07-kim-ai	72	62	more	more	RBR
07-kim-ai	72	63	convincingly	convincingly	RB
07-kim-ai	72	64	due	due	JJ
07-kim-ai	72	65	to	to	IN
07-kim-ai	72	66	the	the	DT
07-kim-ai	72	67	high	high	JJ
07-kim-ai	72	68	stakes	stake	NNS
07-kim-ai	72	69	in-	in-	JJ
07-kim-ai	72	70	volved	volved	NN
07-kim-ai	72	71	.	.	.
07-kim-ai	73	1	Suppose	suppose	VB
07-kim-ai	73	2	that	that	IN
07-kim-ai	73	3	some	some	DT
07-kim-ai	73	4	engineers	engineer	NNS
07-kim-ai	73	5	,	,	,
07-kim-ai	73	6	following	follow	VBG
07-kim-ai	73	7	utilitarianism	utilitarianism	NN
07-kim-ai	73	8	and	and	CC
07-kim-ai	73	9	interpreting	interpret	VBG
07-kim-ai	73	10	victory	victory	NN
07-kim-ai	73	11	as	as	IN
07-kim-ai	73	12	the	the	DT
07-kim-ai	73	13	ul-	ul-	JJ
07-kim-ai	73	14	timate	timate	NN
07-kim-ai	73	15	good	good	JJ
07-kim-ai	73	16	/	/	SYM
07-kim-ai	73	17	utility	utility	NN
07-kim-ai	73	18	,	,	,
07-kim-ai	73	19	wish	wish	VBP
07-kim-ai	73	20	to	to	TO
07-kim-ai	73	21	program	program	VB
07-kim-ai	73	22	an	an	DT
07-kim-ai	73	23	unmanned	unmanned	JJ
07-kim-ai	73	24	aerial	aerial	JJ
07-kim-ai	73	25	vehicle	vehicle	NN
07-kim-ai	73	26	(	(	-LRB-
07-kim-ai	73	27	UAV	UAV	NNP
07-kim-ai	73	28	)	)	-RRB-
07-kim-ai	73	29	to	to	TO
07-kim-ai	73	30	autonomously	autonomously	RB
07-kim-ai	73	31	drop	drop	VB
07-kim-ai	73	32	7For	7for	CD
07-kim-ai	73	33	an	an	DT
07-kim-ai	73	34	overview	overview	NN
07-kim-ai	73	35	,	,	,
07-kim-ai	73	36	see	see	VB
07-kim-ai	73	37	(	(	-LRB-
07-kim-ai	73	38	Sinnott	Sinnott	NNP
07-kim-ai	73	39	-	-	HYPH
07-kim-ai	73	40	Armstrong	Armstrong	NNP
07-kim-ai	73	41	,	,	,
07-kim-ai	73	42	2019	2019	CD
07-kim-ai	73	43	)	)	-RRB-
07-kim-ai	73	44	and	and	CC
07-kim-ai	73	45	(	(	-LRB-
07-kim-ai	73	46	Alexander	Alexander	NNP
07-kim-ai	73	47	and	and	CC
07-kim-ai	73	48	Moore	Moore	NNP
07-kim-ai	73	49	,	,	,
07-kim-ai	73	50	2016	2016	CD
07-kim-ai	73	51	)	)	-RRB-
07-kim-ai	73	52	.	.	.
07-kim-ai	74	1	8For	8for	LS
07-kim-ai	74	2	an	an	DT
07-kim-ai	74	3	empirical	empirical	JJ
07-kim-ai	74	4	study	study	NN
07-kim-ai	74	5	on	on	IN
07-kim-ai	74	6	this	this	DT
07-kim-ai	74	7	,	,	,
07-kim-ai	74	8	see	see	VB
07-kim-ai	74	9	(	(	-LRB-
07-kim-ai	74	10	Cushman	Cushman	NNP
07-kim-ai	74	11	,	,	,
07-kim-ai	74	12	Young	Young	NNP
07-kim-ai	74	13	,	,	,
07-kim-ai	74	14	and	and	CC
07-kim-ai	74	15	Hauser	Hauser	NNP
07-kim-ai	74	16	2006	2006	CD
07-kim-ai	74	17	)	)	-RRB-
07-kim-ai	74	18	.	.	.
07-kim-ai	75	1	For	for	IN
07-kim-ai	75	2	the	the	DT
07-kim-ai	75	3	results	result	NNS
07-kim-ai	75	4	of	of	IN
07-kim-ai	75	5	a	a	DT
07-kim-ai	75	6	similar	similar	JJ
07-kim-ai	75	7	survey	survey	NN
07-kim-ai	75	8	that	that	WDT
07-kim-ai	75	9	involves	involve	VBZ
07-kim-ai	75	10	an	an	DT
07-kim-ai	75	11	autonomous	autonomous	JJ
07-kim-ai	75	12	car	car	NN
07-kim-ai	75	13	instead	instead	RB
07-kim-ai	75	14	of	of	IN
07-kim-ai	75	15	a	a	DT
07-kim-ai	75	16	trolley	trolley	NN
07-kim-ai	75	17	,	,	,
07-kim-ai	75	18	see	see	VB
07-kim-ai	75	19	(	(	-LRB-
07-kim-ai	75	20	Bonnefon	Bonnefon	NNP
07-kim-ai	75	21	,	,	,
07-kim-ai	75	22	Shariff	Shariff	NNP
07-kim-ai	75	23	,	,	,
07-kim-ai	75	24	and	and	CC
07-kim-ai	75	25	Rahwan	Rahwan	NNP
07-kim-ai	75	26	2016	2016	CD
07-kim-ai	75	27	)	)	-RRB-
07-kim-ai	75	28	.	.	.
07-kim-ai	76	1	9For	9for	CD
07-kim-ai	76	2	an	an	DT
07-kim-ai	76	3	attempt	attempt	NN
07-kim-ai	76	4	to	to	TO
07-kim-ai	76	5	identify	identify	VB
07-kim-ai	76	6	moral	moral	JJ
07-kim-ai	76	7	principles	principle	NNS
07-kim-ai	76	8	behind	behind	IN
07-kim-ai	76	9	our	-PRON-	PRP$
07-kim-ai	76	10	moral	moral	JJ
07-kim-ai	76	11	intuition	intuition	NN
07-kim-ai	76	12	in	in	IN
07-kim-ai	76	13	different	different	JJ
07-kim-ai	76	14	versions	version	NNS
07-kim-ai	76	15	of	of	IN
07-kim-ai	76	16	the	the	DT
07-kim-ai	76	17	trolley	trolley	NN
07-kim-ai	76	18	problem	problem	NN
07-kim-ai	76	19	and	and	CC
07-kim-ai	76	20	other	other	JJ
07-kim-ai	76	21	similar	similar	JJ
07-kim-ai	76	22	cases	case	NNS
07-kim-ai	76	23	,	,	,
07-kim-ai	76	24	see	see	VB
07-kim-ai	76	25	(	(	-LRB-
07-kim-ai	76	26	Thomson	Thomson	NNP
07-kim-ai	76	27	1976	1976	CD
07-kim-ai	76	28	)	)	-RRB-
07-kim-ai	76	29	.	.	.
07-kim-ai	77	1	10Some	10some	LS
07-kim-ai	77	2	moral	moral	JJ
07-kim-ai	77	3	philosophers	philosopher	NNS
07-kim-ai	77	4	doubt	doubt	VBP
07-kim-ai	77	5	the	the	DT
07-kim-ai	77	6	value	value	NN
07-kim-ai	77	7	of	of	IN
07-kim-ai	77	8	our	-PRON-	PRP$
07-kim-ai	77	9	moral	moral	JJ
07-kim-ai	77	10	intuition	intuition	NN
07-kim-ai	77	11	in	in	IN
07-kim-ai	77	12	constructing	construct	VBG
07-kim-ai	77	13	a	a	DT
07-kim-ai	77	14	moral	moral	JJ
07-kim-ai	77	15	theory	theory	NN
07-kim-ai	77	16	.	.	.
07-kim-ai	78	1	See	see	VB
07-kim-ai	78	2	(	(	-LRB-
07-kim-ai	78	3	Singer	singer	NN
07-kim-ai	78	4	2005	2005	CD
07-kim-ai	78	5	)	)	-RRB-
07-kim-ai	78	6	,	,	,
07-kim-ai	78	7	for	for	IN
07-kim-ai	78	8	example	example	NN
07-kim-ai	78	9	.	.	.
07-kim-ai	79	1	But	but	CC
07-kim-ai	79	2	a	a	DT
07-kim-ai	79	3	moral	moral	JJ
07-kim-ai	79	4	theory	theory	NN
07-kim-ai	79	5	that	that	IN
07-kim-ai	79	6	clashes	clash	NNS
07-kim-ai	79	7	with	with	IN
07-kim-ai	79	8	common	common	JJ
07-kim-ai	79	9	moral	moral	JJ
07-kim-ai	79	10	intuition	intuition	NN
07-kim-ai	79	11	is	be	VBZ
07-kim-ai	79	12	unlikely	unlikely	JJ
07-kim-ai	79	13	to	to	TO
07-kim-ai	79	14	be	be	VB
07-kim-ai	79	15	sought	seek	VBN
07-kim-ai	79	16	out	out	RP
07-kim-ai	79	17	as	as	IN
07-kim-ai	79	18	a	a	DT
07-kim-ai	79	19	guide	guide	NN
07-kim-ai	79	20	to	to	IN
07-kim-ai	79	21	making	make	VBG
07-kim-ai	79	22	an	an	DT
07-kim-ai	79	23	ethical	ethical	JJ
07-kim-ai	79	24	decision	decision	NN
07-kim-ai	79	25	.	.	.
07-kim-ai	80	1	76	76	CD
07-kim-ai	80	2	Machine	Machine	NNP
07-kim-ai	80	3	Learning	Learning	NNP
07-kim-ai	80	4	,	,	,
07-kim-ai	80	5	Libraries	Libraries	NNPS
07-kim-ai	80	6	,	,	,
07-kim-ai	80	7	and	and	CC
07-kim-ai	80	8	Cross	Cross	NNP
07-kim-ai	80	9	-	-	NNP
07-kim-ai	80	10	Disciplinary	Disciplinary	NNP
07-kim-ai	80	11	ResearchǔChapter	ResearchǔChapter	NNP
07-kim-ai	80	12	7	7	CD
07-kim-ai	80	13	bombs	bomb	NNS
07-kim-ai	80	14	in	in	IN
07-kim-ai	80	15	order	order	NN
07-kim-ai	80	16	to	to	TO
07-kim-ai	80	17	maximize	maximize	VB
07-kim-ai	80	18	the	the	DT
07-kim-ai	80	19	chances	chance	NNS
07-kim-ai	80	20	of	of	IN
07-kim-ai	80	21	victory	victory	NN
07-kim-ai	80	22	.	.	.
07-kim-ai	81	1	That	that	DT
07-kim-ai	81	2	may	may	MD
07-kim-ai	81	3	result	result	VB
07-kim-ai	81	4	in	in	IN
07-kim-ai	81	5	sacrificing	sacrifice	VBG
07-kim-ai	81	6	a	a	DT
07-kim-ai	81	7	greater	great	JJR
07-kim-ai	81	8	num-	num-	NN
07-kim-ai	81	9	ber	ber	NN
07-kim-ai	81	10	of	of	IN
07-kim-ai	81	11	civilians	civilian	NNS
07-kim-ai	81	12	than	than	IN
07-kim-ai	81	13	necessary	necessary	JJ
07-kim-ai	81	14	,	,	,
07-kim-ai	81	15	and	and	CC
07-kim-ai	81	16	many	many	JJ
07-kim-ai	81	17	will	will	MD
07-kim-ai	81	18	consider	consider	VB
07-kim-ai	81	19	this	this	DT
07-kim-ai	81	20	to	to	TO
07-kim-ai	81	21	be	be	VB
07-kim-ai	81	22	morally	morally	RB
07-kim-ai	81	23	wrong	wrong	JJ
07-kim-ai	81	24	.	.	.
07-kim-ai	82	1	Now	now	RB
07-kim-ai	82	2	imagine	imagine	VB
07-kim-ai	82	3	different	different	JJ
07-kim-ai	82	4	engineers	engineer	NNS
07-kim-ai	82	5	who	who	WP
07-kim-ai	82	6	,	,	,
07-kim-ai	82	7	adopting	adopt	VBG
07-kim-ai	82	8	deontology	deontology	NN
07-kim-ai	82	9	and	and	CC
07-kim-ai	82	10	following	follow	VBG
07-kim-ai	82	11	the	the	DT
07-kim-ai	82	12	moral	moral	JJ
07-kim-ai	82	13	principle	principle	NN
07-kim-ai	82	14	of	of	IN
07-kim-ai	82	15	not	not	RB
07-kim-ai	82	16	killing	kill	VBG
07-kim-ai	82	17	people	people	NNS
07-kim-ai	82	18	,	,	,
07-kim-ai	82	19	program	program	VB
07-kim-ai	82	20	a	a	DT
07-kim-ai	82	21	UAV	UAV	NNP
07-kim-ai	82	22	to	to	TO
07-kim-ai	82	23	autonomously	autonomously	RB
07-kim-ai	82	24	act	act	VB
07-kim-ai	82	25	in	in	IN
07-kim-ai	82	26	a	a	DT
07-kim-ai	82	27	manner	manner	NN
07-kim-ai	82	28	that	that	WDT
07-kim-ai	82	29	minimizes	minimize	VBZ
07-kim-ai	82	30	casualties	casualty	NNS
07-kim-ai	82	31	.	.	.
07-kim-ai	83	1	This	this	DT
07-kim-ai	83	2	may	may	MD
07-kim-ai	83	3	lead	lead	VB
07-kim-ai	83	4	to	to	IN
07-kim-ai	83	5	defeat	defeat	NN
07-kim-ai	83	6	on	on	IN
07-kim-ai	83	7	the	the	DT
07-kim-ai	83	8	battlefield	battlefield	NN
07-kim-ai	83	9	,	,	,
07-kim-ai	83	10	because	because	IN
07-kim-ai	83	11	minimizing	minimize	VBG
07-kim-ai	83	12	casualties	casualty	NNS
07-kim-ai	83	13	may	may	MD
07-kim-ai	83	14	not	not	RB
07-kim-ai	83	15	be	be	VB
07-kim-ai	83	16	always	always	RB
07-kim-ai	83	17	advantageous	advantageous	JJ
07-kim-ai	83	18	to	to	IN
07-kim-ai	83	19	winning	win	VBG
07-kim-ai	83	20	a	a	DT
07-kim-ai	83	21	war	war	NN
07-kim-ai	83	22	.	.	.
07-kim-ai	84	1	From	from	IN
07-kim-ai	84	2	these	these	DT
07-kim-ai	84	3	examples	example	NNS
07-kim-ai	84	4	,	,	,
07-kim-ai	84	5	we	-PRON-	PRP
07-kim-ai	84	6	can	can	MD
07-kim-ai	84	7	see	see	VB
07-kim-ai	84	8	that	that	IN
07-kim-ai	84	9	philosophical	philosophical	JJ
07-kim-ai	84	10	insights	insight	NNS
07-kim-ai	84	11	from	from	IN
07-kim-ai	84	12	utilitarian-	utilitarian-	JJ
07-kim-ai	84	13	ism	ism	NN
07-kim-ai	84	14	and	and	CC
07-kim-ai	84	15	deontology	deontology	NN
07-kim-ai	84	16	may	may	MD
07-kim-ai	84	17	provide	provide	VB
07-kim-ai	84	18	little	little	JJ
07-kim-ai	84	19	practical	practical	JJ
07-kim-ai	84	20	guidance	guidance	NN
07-kim-ai	84	21	on	on	IN
07-kim-ai	84	22	how	how	WRB
07-kim-ai	84	23	to	to	TO
07-kim-ai	84	24	program	program	VB
07-kim-ai	84	25	autonomous	autonomous	JJ
07-kim-ai	84	26	AI	AI	NNP
07-kim-ai	84	27	systems	system	NNS
07-kim-ai	84	28	to	to	TO
07-kim-ai	84	29	act	act	VB
07-kim-ai	84	30	morally	morally	RB
07-kim-ai	84	31	.	.	.
07-kim-ai	85	1	Ethicists	ethicist	NNS
07-kim-ai	85	2	seek	seek	VBP
07-kim-ai	85	3	abstract	abstract	JJ
07-kim-ai	85	4	principles	principle	NNS
07-kim-ai	85	5	that	that	WDT
07-kim-ai	85	6	can	can	MD
07-kim-ai	85	7	be	be	VB
07-kim-ai	85	8	generalized	generalize	VBN
07-kim-ai	85	9	.	.	.
07-kim-ai	86	1	For	for	IN
07-kim-ai	86	2	this	this	DT
07-kim-ai	86	3	reason	reason	NN
07-kim-ai	86	4	,	,	,
07-kim-ai	86	5	they	-PRON-	PRP
07-kim-ai	86	6	are	be	VBP
07-kim-ai	86	7	interested	interested	JJ
07-kim-ai	86	8	in	in	IN
07-kim-ai	86	9	borderline	borderline	NN
07-kim-ai	86	10	cases	case	NNS
07-kim-ai	86	11	that	that	WDT
07-kim-ai	86	12	reveal	reveal	VBP
07-kim-ai	86	13	subtle	subtle	JJ
07-kim-ai	86	14	differences	difference	NNS
07-kim-ai	86	15	in	in	IN
07-kim-ai	86	16	our	-PRON-	PRP$
07-kim-ai	86	17	moral	moral	JJ
07-kim-ai	86	18	intuition	intuition	NN
07-kim-ai	86	19	and	and	CC
07-kim-ai	86	20	varying	vary	VBG
07-kim-ai	86	21	moral	moral	JJ
07-kim-ai	86	22	theories	theory	NNS
07-kim-ai	86	23	.	.	.
07-kim-ai	87	1	Their	-PRON-	PRP$
07-kim-ai	87	2	goal	goal	NN
07-kim-ai	87	3	is	be	VBZ
07-kim-ai	87	4	to	to	TO
07-kim-ai	87	5	define	define	VB
07-kim-ai	87	6	what	what	WP
07-kim-ai	87	7	is	be	VBZ
07-kim-ai	87	8	moral	moral	JJ
07-kim-ai	87	9	and	and	CC
07-kim-ai	87	10	investigate	investigate	VB
07-kim-ai	87	11	how	how	WRB
07-kim-ai	87	12	moral	moral	JJ
07-kim-ai	87	13	reasoning	reasoning	NN
07-kim-ai	87	14	works	work	NNS
07-kim-ai	87	15	or	or	CC
07-kim-ai	87	16	should	should	MD
07-kim-ai	87	17	work	work	VB
07-kim-ai	87	18	.	.	.
07-kim-ai	88	1	By	by	IN
07-kim-ai	88	2	contrast	contrast	NN
07-kim-ai	88	3	,	,	,
07-kim-ai	88	4	engineers	engineer	NNS
07-kim-ai	88	5	and	and	CC
07-kim-ai	88	6	programmers	programmer	NNS
07-kim-ai	88	7	pursue	pursue	VBP
07-kim-ai	88	8	practical	practical	JJ
07-kim-ai	88	9	solutions	solution	NNS
07-kim-ai	88	10	to	to	IN
07-kim-ai	88	11	real	real	JJ
07-kim-ai	88	12	-	-	HYPH
07-kim-ai	88	13	life	life	NN
07-kim-ai	88	14	problems	problem	NNS
07-kim-ai	88	15	and	and	CC
07-kim-ai	88	16	look	look	VB
07-kim-ai	88	17	for	for	IN
07-kim-ai	88	18	guidelines	guideline	NNS
07-kim-ai	88	19	that	that	WDT
07-kim-ai	88	20	will	will	MD
07-kim-ai	88	21	help	help	VB
07-kim-ai	88	22	with	with	IN
07-kim-ai	88	23	implementing	implement	VBG
07-kim-ai	88	24	those	those	DT
07-kim-ai	88	25	solutions	solution	NNS
07-kim-ai	88	26	.	.	.
07-kim-ai	89	1	Their	-PRON-	PRP$
07-kim-ai	89	2	focus	focus	NN
07-kim-ai	89	3	is	be	VBZ
07-kim-ai	89	4	on	on	IN
07-kim-ai	89	5	creating	create	VBG
07-kim-ai	89	6	a	a	DT
07-kim-ai	89	7	set	set	NN
07-kim-ai	89	8	of	of	IN
07-kim-ai	89	9	constraints	constraint	NNS
07-kim-ai	89	10	and	and	CC
07-kim-ai	89	11	if	if	IN
07-kim-ai	89	12	-	-	HYPH
07-kim-ai	89	13	then	then	RB
07-kim-ai	89	14	statements	statement	NNS
07-kim-ai	89	15	,	,	,
07-kim-ai	89	16	which	which	WDT
07-kim-ai	89	17	will	will	MD
07-kim-ai	89	18	allow	allow	VB
07-kim-ai	89	19	a	a	DT
07-kim-ai	89	20	machine	machine	NN
07-kim-ai	89	21	to	to	TO
07-kim-ai	89	22	identify	identify	VB
07-kim-ai	89	23	and	and	CC
07-kim-ai	89	24	process	process	VB
07-kim-ai	89	25	morally	morally	RB
07-kim-ai	89	26	relevant	relevant	JJ
07-kim-ai	89	27	considerations	consideration	NNS
07-kim-ai	89	28	,	,	,
07-kim-ai	89	29	so	so	IN
07-kim-ai	89	30	that	that	IN
07-kim-ai	89	31	it	-PRON-	PRP
07-kim-ai	89	32	can	can	MD
07-kim-ai	89	33	determine	determine	VB
07-kim-ai	89	34	and	and	CC
07-kim-ai	89	35	execute	execute	VB
07-kim-ai	89	36	an	an	DT
07-kim-ai	89	37	action	action	NN
07-kim-ai	89	38	that	that	WDT
07-kim-ai	89	39	is	be	VBZ
07-kim-ai	89	40	not	not	RB
07-kim-ai	89	41	only	only	RB
07-kim-ai	89	42	rational	rational	JJ
07-kim-ai	89	43	but	but	CC
07-kim-ai	89	44	also	also	RB
07-kim-ai	89	45	ethical	ethical	JJ
07-kim-ai	89	46	in	in	IN
07-kim-ai	89	47	the	the	DT
07-kim-ai	89	48	given	give	VBN
07-kim-ai	89	49	situation.11	situation.11	NNP
07-kim-ai	89	50	On	on	IN
07-kim-ai	89	51	the	the	DT
07-kim-ai	89	52	other	other	JJ
07-kim-ai	89	53	hand	hand	NN
07-kim-ai	89	54	,	,	,
07-kim-ai	89	55	the	the	DT
07-kim-ai	89	56	goal	goal	NN
07-kim-ai	89	57	of	of	IN
07-kim-ai	89	58	military	military	JJ
07-kim-ai	89	59	commanders	commander	NNS
07-kim-ai	89	60	and	and	CC
07-kim-ai	89	61	soldiers	soldier	NNS
07-kim-ai	89	62	is	be	VBZ
07-kim-ai	89	63	to	to	TO
07-kim-ai	89	64	end	end	VB
07-kim-ai	89	65	a	a	DT
07-kim-ai	89	66	conflict	conflict	NN
07-kim-ai	89	67	,	,	,
07-kim-ai	89	68	bring	bring	VB
07-kim-ai	89	69	peace	peace	NN
07-kim-ai	89	70	,	,	,
07-kim-ai	89	71	and	and	CC
07-kim-ai	89	72	facilitate	facilitate	VB
07-kim-ai	89	73	restoring	restore	VBG
07-kim-ai	89	74	and	and	CC
07-kim-ai	89	75	establishing	establish	VBG
07-kim-ai	89	76	universally	universally	RB
07-kim-ai	89	77	recognized	recognize	VBN
07-kim-ai	89	78	human	human	JJ
07-kim-ai	89	79	values	value	NNS
07-kim-ai	89	80	such	such	JJ
07-kim-ai	89	81	as	as	IN
07-kim-ai	89	82	free-	free-	NNP
07-kim-ai	89	83	dom	dom	NN
07-kim-ai	89	84	,	,	,
07-kim-ai	89	85	equality	equality	NN
07-kim-ai	89	86	,	,	,
07-kim-ai	89	87	justice	justice	NN
07-kim-ai	89	88	,	,	,
07-kim-ai	89	89	and	and	CC
07-kim-ai	89	90	self	self	NN
07-kim-ai	89	91	-	-	HYPH
07-kim-ai	89	92	determination	determination	NN
07-kim-ai	89	93	.	.	.
07-kim-ai	90	1	In	in	IN
07-kim-ai	90	2	order	order	NN
07-kim-ai	90	3	to	to	TO
07-kim-ai	90	4	achieve	achieve	VB
07-kim-ai	90	5	this	this	DT
07-kim-ai	90	6	goal	goal	NN
07-kim-ai	90	7	,	,	,
07-kim-ai	90	8	they	-PRON-	PRP
07-kim-ai	90	9	must	must	MD
07-kim-ai	90	10	make	make	VB
07-kim-ai	90	11	the	the	DT
07-kim-ai	90	12	best	good	JJS
07-kim-ai	90	13	strategic	strategic	JJ
07-kim-ai	90	14	decisions	decision	NNS
07-kim-ai	90	15	and	and	CC
07-kim-ai	90	16	take	take	VB
07-kim-ai	90	17	the	the	DT
07-kim-ai	90	18	most	most	RBS
07-kim-ai	90	19	appropriate	appropriate	JJ
07-kim-ai	90	20	actions	action	NNS
07-kim-ai	90	21	.	.	.
07-kim-ai	91	1	In	in	IN
07-kim-ai	91	2	deciding	decide	VBG
07-kim-ai	91	3	on	on	IN
07-kim-ai	91	4	those	those	DT
07-kim-ai	91	5	actions	action	NNS
07-kim-ai	91	6	,	,	,
07-kim-ai	91	7	they	-PRON-	PRP
07-kim-ai	91	8	are	be	VBP
07-kim-ai	91	9	also	also	RB
07-kim-ai	91	10	responsible	responsible	JJ
07-kim-ai	91	11	for	for	IN
07-kim-ai	91	12	abiding	abide	VBG
07-kim-ai	91	13	by	by	IN
07-kim-ai	91	14	the	the	DT
07-kim-ai	91	15	principles	principle	NNS
07-kim-ai	91	16	of	of	IN
07-kim-ai	91	17	jus	jus	NN
07-kim-ai	91	18	in	in	IN
07-kim-ai	91	19	bello	bello	NN
07-kim-ai	91	20	and	and	CC
07-kim-ai	91	21	for	for	IN
07-kim-ai	91	22	not	not	RB
07-kim-ai	91	23	abdicating	abdicate	VBG
07-kim-ai	91	24	their	-PRON-	PRP$
07-kim-ai	91	25	moral	moral	JJ
07-kim-ai	91	26	responsibility	responsibility	NN
07-kim-ai	91	27	,	,	,
07-kim-ai	91	28	protecting	protect	VBG
07-kim-ai	91	29	civilians	civilian	NNS
07-kim-ai	91	30	and	and	CC
07-kim-ai	91	31	minimizing	minimize	VBG
07-kim-ai	91	32	harm	harm	NN
07-kim-ai	91	33	,	,	,
07-kim-ai	91	34	violence	violence	NN
07-kim-ai	91	35	,	,	,
07-kim-ai	91	36	and	and	CC
07-kim-ai	91	37	destruction	destruction	NN
07-kim-ai	91	38	as	as	RB
07-kim-ai	91	39	much	much	RB
07-kim-ai	91	40	as	as	IN
07-kim-ai	91	41	possible.12	possible.12	NNP
07-kim-ai	91	42	The	the	DT
07-kim-ai	91	43	goal	goal	NN
07-kim-ai	91	44	of	of	IN
07-kim-ai	91	45	military	military	JJ
07-kim-ai	91	46	commanders	commander	NNS
07-kim-ai	91	47	and	and	CC
07-kim-ai	91	48	soldiers	soldier	NNS
07-kim-ai	91	49	,	,	,
07-kim-ai	91	50	therefore	therefore	RB
07-kim-ai	91	51	,	,	,
07-kim-ai	91	52	differs	differ	NNS
07-kim-ai	91	53	from	from	IN
07-kim-ai	91	54	those	those	DT
07-kim-ai	91	55	of	of	IN
07-kim-ai	91	56	moral	moral	JJ
07-kim-ai	91	57	philosophers	philosopher	NNS
07-kim-ai	91	58	or	or	CC
07-kim-ai	91	59	of	of	IN
07-kim-ai	91	60	the	the	DT
07-kim-ai	91	61	engineers	engineer	NNS
07-kim-ai	91	62	who	who	WP
07-kim-ai	91	63	build	build	VBP
07-kim-ai	91	64	autonomous	autonomous	JJ
07-kim-ai	91	65	weapons	weapon	NNS
07-kim-ai	91	66	.	.	.
07-kim-ai	92	1	They	-PRON-	PRP
07-kim-ai	92	2	are	be	VBP
07-kim-ai	92	3	obligated	obligate	VBN
07-kim-ai	92	4	to	to	TO
07-kim-ai	92	5	make	make	VB
07-kim-ai	92	6	quick	quick	JJ
07-kim-ai	92	7	decisions	decision	NNS
07-kim-ai	92	8	in	in	IN
07-kim-ai	92	9	a	a	DT
07-kim-ai	92	10	life	life	NN
07-kim-ai	92	11	-	-	HYPH
07-kim-ai	92	12	or	or	CC
07-kim-ai	92	13	-	-	HYPH
07-kim-ai	92	14	death	death	NN
07-kim-ai	92	15	situation	situation	NN
07-kim-ai	92	16	while	while	IN
07-kim-ai	92	17	working	work	VBG
07-kim-ai	92	18	with	with	IN
07-kim-ai	92	19	AI	AI	NNP
07-kim-ai	92	20	-	-	HYPH
07-kim-ai	92	21	powered	power	VBN
07-kim-ai	92	22	military	military	JJ
07-kim-ai	92	23	systems	system	NNS
07-kim-ai	92	24	.	.	.
07-kim-ai	93	1	These	these	DT
07-kim-ai	93	2	different	different	JJ
07-kim-ai	93	3	goals	goal	NNS
07-kim-ai	93	4	and	and	CC
07-kim-ai	93	5	interests	interest	NNS
07-kim-ai	93	6	explain	explain	VBP
07-kim-ai	93	7	why	why	WRB
07-kim-ai	93	8	moral	moral	JJ
07-kim-ai	93	9	philosophers	philosopher	NNS
07-kim-ai	93	10	’	’	POS
07-kim-ai	93	11	discussion	discussion	NN
07-kim-ai	93	12	on	on	IN
07-kim-ai	93	13	the	the	DT
07-kim-ai	93	14	trolley	trolley	NN
07-kim-ai	93	15	problem	problem	NN
07-kim-ai	93	16	may	may	MD
07-kim-ai	93	17	be	be	VB
07-kim-ai	93	18	disappointing	disappointing	JJ
07-kim-ai	93	19	to	to	IN
07-kim-ai	93	20	AI	AI	NNP
07-kim-ai	93	21	programmers	programmer	NNS
07-kim-ai	93	22	or	or	CC
07-kim-ai	93	23	military	military	JJ
07-kim-ai	93	24	commanders	commander	NNS
07-kim-ai	93	25	and	and	CC
07-kim-ai	93	26	soldiers	soldier	NNS
07-kim-ai	93	27	.	.	.
07-kim-ai	94	1	Ethics	ethic	NNS
07-kim-ai	94	2	does	do	VBZ
07-kim-ai	94	3	not	not	RB
07-kim-ai	94	4	provide	provide	VB
07-kim-ai	94	5	an	an	DT
07-kim-ai	94	6	easy	easy	JJ
07-kim-ai	94	7	answer	answer	NN
07-kim-ai	94	8	to	to	IN
07-kim-ai	94	9	the	the	DT
07-kim-ai	94	10	question	question	NN
07-kim-ai	94	11	of	of	IN
07-kim-ai	94	12	how	how	WRB
07-kim-ai	94	13	one	one	PRP
07-kim-ai	94	14	should	should	MD
07-kim-ai	94	15	program	program	VB
07-kim-ai	94	16	moral	moral	JJ
07-kim-ai	94	17	decision-	decision-	NN
07-kim-ai	94	18	making	make	VBG
07-kim-ai	94	19	into	into	IN
07-kim-ai	94	20	intelligent	intelligent	JJ
07-kim-ai	94	21	machines	machine	NNS
07-kim-ai	94	22	.	.	.
07-kim-ai	95	1	Nor	nor	CC
07-kim-ai	95	2	does	do	VBZ
07-kim-ai	95	3	it	-PRON-	PRP
07-kim-ai	95	4	prescribe	prescribe	VB
07-kim-ai	95	5	the	the	DT
07-kim-ai	95	6	right	right	JJ
07-kim-ai	95	7	moral	moral	JJ
07-kim-ai	95	8	decision	decision	NN
07-kim-ai	95	9	in	in	IN
07-kim-ai	95	10	a	a	DT
07-kim-ai	95	11	battlefield	battlefield	NN
07-kim-ai	95	12	.	.	.
07-kim-ai	96	1	But	but	CC
07-kim-ai	96	2	taking	take	VBG
07-kim-ai	96	3	this	this	DT
07-kim-ai	96	4	as	as	IN
07-kim-ai	96	5	a	a	DT
07-kim-ai	96	6	shortcoming	shortcoming	NN
07-kim-ai	96	7	of	of	IN
07-kim-ai	96	8	ethics	ethic	NNS
07-kim-ai	96	9	is	be	VBZ
07-kim-ai	96	10	missing	miss	VBG
07-kim-ai	96	11	the	the	DT
07-kim-ai	96	12	point	point	NN
07-kim-ai	96	13	.	.	.
07-kim-ai	97	1	The	the	DT
07-kim-ai	97	2	role	role	NN
07-kim-ai	97	3	of	of	IN
07-kim-ai	97	4	moral	moral	JJ
07-kim-ai	97	5	philosophy	philosophy	NN
07-kim-ai	97	6	is	be	VBZ
07-kim-ai	97	7	not	not	RB
07-kim-ai	97	8	to	to	TO
07-kim-ai	97	9	make	make	VB
07-kim-ai	97	10	decision	decision	NN
07-kim-ai	97	11	-	-	HYPH
07-kim-ai	97	12	making	make	VBG
07-kim-ai	97	13	easier	easy	JJR
07-kim-ai	97	14	but	but	CC
07-kim-ai	97	15	to	to	TO
07-kim-ai	97	16	highlight	highlight	VB
07-kim-ai	97	17	and	and	CC
07-kim-ai	97	18	articulate	articulate	VB
07-kim-ai	97	19	the	the	DT
07-kim-ai	97	20	difficulty	difficulty	NN
07-kim-ai	97	21	and	and	CC
07-kim-ai	97	22	complexity	complexity	NN
07-kim-ai	97	23	involved	involve	VBN
07-kim-ai	97	24	in	in	IN
07-kim-ai	97	25	it	-PRON-	PRP
07-kim-ai	97	26	.	.	.
07-kim-ai	98	1	Ethical	ethical	JJ
07-kim-ai	98	2	Challenges	Challenges	NNPS
07-kim-ai	98	3	from	from	IN
07-kim-ai	98	4	Autonomous	Autonomous	NNP
07-kim-ai	98	5	AI	AI	NNP
07-kim-ai	98	6	Systems	Systems	NNP
07-kim-ai	98	7	The	the	DT
07-kim-ai	98	8	complexity	complexity	NN
07-kim-ai	98	9	of	of	IN
07-kim-ai	98	10	ethical	ethical	JJ
07-kim-ai	98	11	questions	question	NNS
07-kim-ai	98	12	means	mean	VBZ
07-kim-ai	98	13	that	that	IN
07-kim-ai	98	14	dealing	deal	VBG
07-kim-ai	98	15	with	with	IN
07-kim-ai	98	16	the	the	DT
07-kim-ai	98	17	morality	morality	NN
07-kim-ai	98	18	of	of	IN
07-kim-ai	98	19	an	an	DT
07-kim-ai	98	20	action	action	NN
07-kim-ai	98	21	by	by	IN
07-kim-ai	98	22	an	an	DT
07-kim-ai	98	23	autonomous	autonomous	JJ
07-kim-ai	98	24	AI	ai	NN
07-kim-ai	98	25	system	system	NN
07-kim-ai	98	26	will	will	MD
07-kim-ai	98	27	require	require	VB
07-kim-ai	98	28	more	more	JJR
07-kim-ai	98	29	than	than	IN
07-kim-ai	98	30	a	a	DT
07-kim-ai	98	31	clever	clever	JJ
07-kim-ai	98	32	engineering	engineering	NN
07-kim-ai	98	33	or	or	CC
07-kim-ai	98	34	programming	programming	NN
07-kim-ai	98	35	solution	solution	NN
07-kim-ai	98	36	.	.	.
07-kim-ai	99	1	The	the	DT
07-kim-ai	99	2	fact	fact	NN
07-kim-ai	99	3	that	that	IN
07-kim-ai	99	4	ethics	ethic	NNS
07-kim-ai	99	5	does	do	VBZ
07-kim-ai	99	6	not	not	RB
07-kim-ai	99	7	eliminate	eliminate	VB
07-kim-ai	99	8	the	the	DT
07-kim-ai	99	9	inherent	inherent	JJ
07-kim-ai	99	10	ambiguity	ambiguity	NN
07-kim-ai	99	11	in	in	IN
07-kim-ai	99	12	many	many	JJ
07-kim-ai	99	13	moral	moral	JJ
07-kim-ai	99	14	decisions	decision	NNS
07-kim-ai	99	15	should	should	MD
07-kim-ai	99	16	not	not	RB
07-kim-ai	99	17	lead	lead	VB
07-kim-ai	99	18	to	to	IN
07-kim-ai	99	19	the	the	DT
07-kim-ai	99	20	dismissal	dismissal	NN
07-kim-ai	99	21	of	of	IN
07-kim-ai	99	22	ethical	ethical	JJ
07-kim-ai	99	23	challenges	challenge	NNS
07-kim-ai	99	24	from	from	IN
07-kim-ai	99	25	autonomous	autonomous	JJ
07-kim-ai	99	26	AI	AI	NNP
07-kim-ai	99	27	systems	system	NNS
07-kim-ai	99	28	.	.	.
07-kim-ai	100	1	By	by	IN
07-kim-ai	100	2	injecting	inject	VBG
07-kim-ai	100	3	the	the	DT
07-kim-ai	100	4	capacity	capacity	NN
07-kim-ai	100	5	for	for	IN
07-kim-ai	100	6	autonomous	autonomous	JJ
07-kim-ai	100	7	decision	decision	NN
07-kim-ai	100	8	-	-	HYPH
07-kim-ai	100	9	making	making	NN
07-kim-ai	100	10	into	into	IN
07-kim-ai	100	11	machines	machine	NNS
07-kim-ai	100	12	,	,	,
07-kim-ai	100	13	AI	AI	NNP
07-kim-ai	100	14	can	can	MD
07-kim-ai	100	15	fundamentally	fundamentally	RB
07-kim-ai	100	16	transform	transform	VB
07-kim-ai	100	17	any	any	DT
07-kim-ai	100	18	given	give	VBN
07-kim-ai	100	19	field	field	NN
07-kim-ai	100	20	.	.	.
07-kim-ai	101	1	For	for	IN
07-kim-ai	101	2	example	example	NN
07-kim-ai	101	3	,	,	,
07-kim-ai	101	4	AI	AI	NNP
07-kim-ai	101	5	-	-	HYPH
07-kim-ai	101	6	powered	power	VBN
07-kim-ai	101	7	military	military	JJ
07-kim-ai	101	8	robots	robot	NNS
07-kim-ai	101	9	are	be	VBP
07-kim-ai	101	10	not	not	RB
07-kim-ai	101	11	just	just	RB
07-kim-ai	101	12	another	another	DT
07-kim-ai	101	13	kind	kind	NN
07-kim-ai	101	14	of	of	IN
07-kim-ai	101	15	weapon	weapon	NN
07-kim-ai	101	16	.	.	.
07-kim-ai	102	1	When	when	WRB
07-kim-ai	102	2	widely	widely	RB
07-kim-ai	102	3	deployed	deploy	VBN
07-kim-ai	102	4	,	,	,
07-kim-ai	102	5	they	-PRON-	PRP
07-kim-ai	102	6	can	can	MD
07-kim-ai	102	7	change	change	VB
07-kim-ai	102	8	the	the	DT
07-kim-ai	102	9	nature	nature	NN
07-kim-ai	102	10	of	of	IN
07-kim-ai	102	11	war	war	NN
07-kim-ai	102	12	itself	-PRON-	PRP
07-kim-ai	102	13	.	.	.
07-kim-ai	103	1	Described	describe	VBN
07-kim-ai	103	2	below	below	IN
07-kim-ai	103	3	are	be	VBP
07-kim-ai	103	4	some	some	DT
07-kim-ai	103	5	of	of	IN
07-kim-ai	103	6	the	the	DT
07-kim-ai	103	7	signif-	signif-	JJ
07-kim-ai	103	8	icant	icant	NNP
07-kim-ai	103	9	ethical	ethical	JJ
07-kim-ai	103	10	challenges	challenge	NNS
07-kim-ai	103	11	that	that	WDT
07-kim-ai	103	12	autonomous	autonomous	JJ
07-kim-ai	103	13	AI	AI	NNP
07-kim-ai	103	14	systems	system	NNS
07-kim-ai	103	15	such	such	JJ
07-kim-ai	103	16	as	as	IN
07-kim-ai	103	17	military	military	JJ
07-kim-ai	103	18	robots	robot	NNS
07-kim-ai	103	19	present	present	JJ
07-kim-ai	103	20	.	.	.
07-kim-ai	104	1	Note	note	VB
07-kim-ai	104	2	that	that	IN
07-kim-ai	104	3	11Note	11note	CD
07-kim-ai	104	4	that	that	IN
07-kim-ai	104	5	this	this	DT
07-kim-ai	104	6	moral	moral	JJ
07-kim-ai	104	7	decision	decision	NN
07-kim-ai	104	8	-	-	HYPH
07-kim-ai	104	9	making	make	VBG
07-kim-ai	104	10	process	process	NN
07-kim-ai	104	11	can	can	MD
07-kim-ai	104	12	be	be	VB
07-kim-ai	104	13	modeled	model	VBN
07-kim-ai	104	14	with	with	IN
07-kim-ai	104	15	a	a	DT
07-kim-ai	104	16	rule	rule	NN
07-kim-ai	104	17	-	-	HYPH
07-kim-ai	104	18	based	base	VBN
07-kim-ai	104	19	symbolic	symbolic	JJ
07-kim-ai	104	20	AI	AI	NNP
07-kim-ai	104	21	approach	approach	NN
07-kim-ai	104	22	,	,	,
07-kim-ai	104	23	a	a	DT
07-kim-ai	104	24	machine	machine	NN
07-kim-ai	104	25	learning	learn	VBG
07-kim-ai	104	26	approach	approach	NN
07-kim-ai	104	27	,	,	,
07-kim-ai	104	28	or	or	CC
07-kim-ai	104	29	a	a	DT
07-kim-ai	104	30	combination	combination	NN
07-kim-ai	104	31	of	of	IN
07-kim-ai	104	32	both	both	DT
07-kim-ai	104	33	.	.	.
07-kim-ai	105	1	See	see	VB
07-kim-ai	105	2	Vincent	Vincent	NNP
07-kim-ai	105	3	Conitzer	Conitzer	NNP
07-kim-ai	105	4	et	et	NNP
07-kim-ai	105	5	al	al	NNP
07-kim-ai	105	6	.	.	.
07-kim-ai	106	1	2017	2017	CD
07-kim-ai	106	2	.	.	.
07-kim-ai	107	1	12For	12for	CD
07-kim-ai	107	2	the	the	DT
07-kim-ai	107	3	principles	principle	NNS
07-kim-ai	107	4	of	of	IN
07-kim-ai	107	5	jus	jus	NN
07-kim-ai	107	6	in	in	IN
07-kim-ai	107	7	bello	bello	NN
07-kim-ai	107	8	,	,	,
07-kim-ai	107	9	see	see	VB
07-kim-ai	107	10	International	International	NNP
07-kim-ai	107	11	Committee	Committee	NNP
07-kim-ai	107	12	of	of	IN
07-kim-ai	107	13	the	the	DT
07-kim-ai	107	14	Red	Red	NNP
07-kim-ai	107	15	Cross	Cross	NNP
07-kim-ai	107	16	2015	2015	CD
07-kim-ai	107	17	.	.	.
07-kim-ai	108	1	Kim	Kim	NNP
07-kim-ai	108	2	77	77	CD
07-kim-ai	108	3	in	in	IN
07-kim-ai	108	4	spite	spite	NN
07-kim-ai	108	5	of	of	IN
07-kim-ai	108	6	these	these	DT
07-kim-ai	108	7	ethical	ethical	JJ
07-kim-ai	108	8	concerns	concern	NNS
07-kim-ai	108	9	,	,	,
07-kim-ai	108	10	autonomous	autonomous	JJ
07-kim-ai	108	11	AI	AI	NNP
07-kim-ai	108	12	systems	system	NNS
07-kim-ai	108	13	are	be	VBP
07-kim-ai	108	14	likely	likely	JJ
07-kim-ai	108	15	to	to	TO
07-kim-ai	108	16	continue	continue	VB
07-kim-ai	108	17	to	to	TO
07-kim-ai	108	18	be	be	VB
07-kim-ai	108	19	developed	develop	VBN
07-kim-ai	108	20	and	and	CC
07-kim-ai	108	21	adopted	adopt	VBN
07-kim-ai	108	22	in	in	IN
07-kim-ai	108	23	many	many	JJ
07-kim-ai	108	24	areas	area	NNS
07-kim-ai	108	25	as	as	IN
07-kim-ai	108	26	a	a	DT
07-kim-ai	108	27	way	way	NN
07-kim-ai	108	28	to	to	TO
07-kim-ai	108	29	increase	increase	VB
07-kim-ai	108	30	efficiency	efficiency	NN
07-kim-ai	108	31	and	and	CC
07-kim-ai	108	32	lower	low	JJR
07-kim-ai	108	33	cost	cost	NN
07-kim-ai	108	34	.	.	.
07-kim-ai	109	1	(	(	-LRB-
07-kim-ai	109	2	a	a	LS
07-kim-ai	109	3	)	)	-RRB-
07-kim-ai	109	4	Moral	moral	JJ
07-kim-ai	109	5	desensitization	desensitization	NN
07-kim-ai	109	6	AI	ai	NN
07-kim-ai	109	7	-	-	HYPH
07-kim-ai	109	8	powered	power	VBN
07-kim-ai	109	9	military	military	JJ
07-kim-ai	109	10	robots	robot	NNS
07-kim-ai	109	11	are	be	VBP
07-kim-ai	109	12	more	more	RBR
07-kim-ai	109	13	capable	capable	JJ
07-kim-ai	109	14	than	than	IN
07-kim-ai	109	15	merely	merely	RB
07-kim-ai	109	16	remotely	remotely	RB
07-kim-ai	109	17	-	-	HYPH
07-kim-ai	109	18	operated	operate	VBN
07-kim-ai	109	19	weapons	weapon	NNS
07-kim-ai	109	20	.	.	.
07-kim-ai	110	1	They	-PRON-	PRP
07-kim-ai	110	2	can	can	MD
07-kim-ai	110	3	identify	identify	VB
07-kim-ai	110	4	a	a	DT
07-kim-ai	110	5	target	target	NN
07-kim-ai	110	6	and	and	CC
07-kim-ai	110	7	initiate	initiate	VB
07-kim-ai	110	8	an	an	DT
07-kim-ai	110	9	attack	attack	NN
07-kim-ai	110	10	on	on	IN
07-kim-ai	110	11	their	-PRON-	PRP$
07-kim-ai	110	12	own	own	JJ
07-kim-ai	110	13	.	.	.
07-kim-ai	111	1	Due	due	IN
07-kim-ai	111	2	to	to	IN
07-kim-ai	111	3	their	-PRON-	PRP$
07-kim-ai	111	4	autonomy	autonomy	NN
07-kim-ai	111	5	,	,	,
07-kim-ai	111	6	military	military	JJ
07-kim-ai	111	7	robots	robot	NNS
07-kim-ai	111	8	can	can	MD
07-kim-ai	111	9	significantly	significantly	RB
07-kim-ai	111	10	increase	increase	VB
07-kim-ai	111	11	the	the	DT
07-kim-ai	111	12	distance	distance	NN
07-kim-ai	111	13	between	between	IN
07-kim-ai	111	14	the	the	DT
07-kim-ai	111	15	party	party	NN
07-kim-ai	111	16	that	that	WDT
07-kim-ai	111	17	kills	kill	VBZ
07-kim-ai	111	18	and	and	CC
07-kim-ai	111	19	the	the	DT
07-kim-ai	111	20	party	party	NN
07-kim-ai	111	21	that	that	WDT
07-kim-ai	111	22	gets	get	VBZ
07-kim-ai	111	23	killed	kill	VBN
07-kim-ai	111	24	(	(	-LRB-
07-kim-ai	111	25	Sharkey	Sharkey	NNP
07-kim-ai	111	26	2012	2012	CD
07-kim-ai	111	27	)	)	-RRB-
07-kim-ai	111	28	.	.	.
07-kim-ai	112	1	This	this	DT
07-kim-ai	112	2	increase	increase	NN
07-kim-ai	112	3	,	,	,
07-kim-ai	112	4	however	however	RB
07-kim-ai	112	5	,	,	,
07-kim-ai	112	6	may	may	MD
07-kim-ai	112	7	lead	lead	VB
07-kim-ai	112	8	people	people	NNS
07-kim-ai	112	9	to	to	TO
07-kim-ai	112	10	surrender	surrender	VB
07-kim-ai	112	11	their	-PRON-	PRP$
07-kim-ai	112	12	own	own	JJ
07-kim-ai	112	13	moral	moral	JJ
07-kim-ai	112	14	responsi-	responsi-	CC
07-kim-ai	112	15	bility	bility	NN
07-kim-ai	112	16	to	to	IN
07-kim-ai	112	17	a	a	DT
07-kim-ai	112	18	machine	machine	NN
07-kim-ai	112	19	,	,	,
07-kim-ai	112	20	thereby	thereby	RB
07-kim-ai	112	21	resulting	result	VBG
07-kim-ai	112	22	in	in	IN
07-kim-ai	112	23	the	the	DT
07-kim-ai	112	24	loss	loss	NN
07-kim-ai	112	25	of	of	IN
07-kim-ai	112	26	humanity	humanity	NN
07-kim-ai	112	27	,	,	,
07-kim-ai	112	28	which	which	WDT
07-kim-ai	112	29	is	be	VBZ
07-kim-ai	112	30	a	a	DT
07-kim-ai	112	31	serious	serious	JJ
07-kim-ai	112	32	moral	moral	JJ
07-kim-ai	112	33	risk	risk	NN
07-kim-ai	112	34	(	(	-LRB-
07-kim-ai	112	35	Davis	Davis	NNP
07-kim-ai	112	36	2007	2007	CD
07-kim-ai	112	37	)	)	-RRB-
07-kim-ai	112	38	.	.	.
07-kim-ai	113	1	The	the	DT
07-kim-ai	113	2	more	more	RBR
07-kim-ai	113	3	autonomous	autonomous	JJ
07-kim-ai	113	4	military	military	JJ
07-kim-ai	113	5	robots	robot	NNS
07-kim-ai	113	6	become	become	VBP
07-kim-ai	113	7	,	,	,
07-kim-ai	113	8	the	the	DT
07-kim-ai	113	9	less	less	JJR
07-kim-ai	113	10	responsibility	responsibility	NN
07-kim-ai	113	11	humans	human	NNS
07-kim-ai	113	12	will	will	MD
07-kim-ai	113	13	feel	feel	VB
07-kim-ai	113	14	regarding	regard	VBG
07-kim-ai	113	15	their	-PRON-	PRP$
07-kim-ai	113	16	life	life	NN
07-kim-ai	113	17	-	-	HYPH
07-kim-ai	113	18	or	or	CC
07-kim-ai	113	19	-	-	HYPH
07-kim-ai	113	20	death	death	NN
07-kim-ai	113	21	decisions	decision	NNS
07-kim-ai	113	22	.	.	.
07-kim-ai	114	1	(	(	-LRB-
07-kim-ai	114	2	b	b	NN
07-kim-ai	114	3	)	)	-RRB-
07-kim-ai	114	4	Unintended	unintended	JJ
07-kim-ai	114	5	outcome	outcome	NN
07-kim-ai	114	6	The	the	DT
07-kim-ai	114	7	side	side	NN
07-kim-ai	114	8	that	that	WDT
07-kim-ai	114	9	deploys	deploy	VBZ
07-kim-ai	114	10	AI	ai	NN
07-kim-ai	114	11	-	-	HYPH
07-kim-ai	114	12	powered	power	VBN
07-kim-ai	114	13	military	military	JJ
07-kim-ai	114	14	robots	robot	NNS
07-kim-ai	114	15	is	be	VBZ
07-kim-ai	114	16	likely	likely	JJ
07-kim-ai	114	17	to	to	TO
07-kim-ai	114	18	suffer	suffer	VB
07-kim-ai	114	19	fewer	few	JJR
07-kim-ai	114	20	casualties	casualty	NNS
07-kim-ai	114	21	itself	-PRON-	PRP
07-kim-ai	114	22	while	while	IN
07-kim-ai	114	23	inflicting	inflict	VBG
07-kim-ai	114	24	more	more	JJR
07-kim-ai	114	25	casualties	casualty	NNS
07-kim-ai	114	26	on	on	IN
07-kim-ai	114	27	the	the	DT
07-kim-ai	114	28	enemy	enemy	NN
07-kim-ai	114	29	side	side	NN
07-kim-ai	114	30	.	.	.
07-kim-ai	115	1	This	this	DT
07-kim-ai	115	2	may	may	MD
07-kim-ai	115	3	make	make	VB
07-kim-ai	115	4	the	the	DT
07-kim-ai	115	5	military	military	NN
07-kim-ai	115	6	more	more	RBR
07-kim-ai	115	7	inclined	inclined	JJ
07-kim-ai	115	8	to	to	TO
07-kim-ai	115	9	start	start	VB
07-kim-ai	115	10	a	a	DT
07-kim-ai	115	11	war	war	NN
07-kim-ai	115	12	.	.	.
07-kim-ai	116	1	Ironically	ironically	RB
07-kim-ai	116	2	,	,	,
07-kim-ai	116	3	when	when	WRB
07-kim-ai	116	4	everyone	everyone	NN
07-kim-ai	116	5	thinks	think	VBZ
07-kim-ai	116	6	and	and	CC
07-kim-ai	116	7	acts	act	VBZ
07-kim-ai	116	8	this	this	DT
07-kim-ai	116	9	way	way	NN
07-kim-ai	116	10	,	,	,
07-kim-ai	116	11	the	the	DT
07-kim-ai	116	12	number	number	NN
07-kim-ai	116	13	of	of	IN
07-kim-ai	116	14	wars	war	NNS
07-kim-ai	116	15	and	and	CC
07-kim-ai	116	16	the	the	DT
07-kim-ai	116	17	overall	overall	JJ
07-kim-ai	116	18	amount	amount	NN
07-kim-ai	116	19	of	of	IN
07-kim-ai	116	20	violence	violence	NN
07-kim-ai	116	21	and	and	CC
07-kim-ai	116	22	destruction	destruction	NN
07-kim-ai	116	23	in	in	IN
07-kim-ai	116	24	the	the	DT
07-kim-ai	116	25	world	world	NN
07-kim-ai	116	26	will	will	MD
07-kim-ai	116	27	only	only	RB
07-kim-ai	116	28	increase.13	increase.13	NNP
07-kim-ai	116	29	(	(	-LRB-
07-kim-ai	116	30	c	c	NN
07-kim-ai	116	31	)	)	-RRB-
07-kim-ai	116	32	Surrender	surrender	NN
07-kim-ai	116	33	of	of	IN
07-kim-ai	116	34	moral	moral	JJ
07-kim-ai	116	35	agency	agency	NN
07-kim-ai	116	36	AI	AI	NNP
07-kim-ai	116	37	-	-	HYPH
07-kim-ai	116	38	powered	power	VBN
07-kim-ai	116	39	military	military	JJ
07-kim-ai	116	40	robots	robot	NNS
07-kim-ai	116	41	may	may	MD
07-kim-ai	116	42	fail	fail	VB
07-kim-ai	116	43	to	to	TO
07-kim-ai	116	44	distinguish	distinguish	VB
07-kim-ai	116	45	innocents	innocent	NNS
07-kim-ai	116	46	from	from	IN
07-kim-ai	116	47	combatants	combatant	NNS
07-kim-ai	116	48	and	and	CC
07-kim-ai	116	49	kill	kill	VB
07-kim-ai	116	50	the	the	DT
07-kim-ai	116	51	for-	for-	NN
07-kim-ai	116	52	mer	mer	NNP
07-kim-ai	116	53	.	.	.
07-kim-ai	117	1	In	in	IN
07-kim-ai	117	2	such	such	PDT
07-kim-ai	117	3	a	a	DT
07-kim-ai	117	4	case	case	NN
07-kim-ai	117	5	,	,	,
07-kim-ai	117	6	can	can	MD
07-kim-ai	117	7	we	-PRON-	PRP
07-kim-ai	117	8	be	be	VB
07-kim-ai	117	9	justified	justify	VBN
07-kim-ai	117	10	in	in	IN
07-kim-ai	117	11	letting	let	VBG
07-kim-ai	117	12	robots	robot	NNS
07-kim-ai	117	13	take	take	VB
07-kim-ai	117	14	the	the	DT
07-kim-ai	117	15	lives	life	NNS
07-kim-ai	117	16	of	of	IN
07-kim-ai	117	17	other	other	JJ
07-kim-ai	117	18	human	human	JJ
07-kim-ai	117	19	beings	being	NNS
07-kim-ai	117	20	?	?	.
07-kim-ai	118	1	Some	some	DT
07-kim-ai	118	2	may	may	MD
07-kim-ai	118	3	argue	argue	VB
07-kim-ai	118	4	that	that	IN
07-kim-ai	118	5	only	only	RB
07-kim-ai	118	6	humans	human	NNS
07-kim-ai	118	7	should	should	MD
07-kim-ai	118	8	decide	decide	VB
07-kim-ai	118	9	to	to	TO
07-kim-ai	118	10	kill	kill	VB
07-kim-ai	118	11	other	other	JJ
07-kim-ai	118	12	humans	human	NNS
07-kim-ai	118	13	,	,	,
07-kim-ai	118	14	not	not	RB
07-kim-ai	118	15	machines	machine	NNS
07-kim-ai	118	16	(	(	-LRB-
07-kim-ai	118	17	Davis	Davis	NNP
07-kim-ai	118	18	2007	2007	CD
07-kim-ai	118	19	)	)	-RRB-
07-kim-ai	118	20	.	.	.
07-kim-ai	119	1	Is	be	VBZ
07-kim-ai	119	2	it	-PRON-	PRP
07-kim-ai	119	3	permissible	permissible	JJ
07-kim-ai	119	4	for	for	IN
07-kim-ai	119	5	people	people	NNS
07-kim-ai	119	6	to	to	TO
07-kim-ai	119	7	delegate	delegate	VB
07-kim-ai	119	8	such	such	PDT
07-kim-ai	119	9	a	a	DT
07-kim-ai	119	10	decision	decision	NN
07-kim-ai	119	11	to	to	IN
07-kim-ai	119	12	AI	AI	NNP
07-kim-ai	119	13	?	?	.
07-kim-ai	120	1	(	(	-LRB-
07-kim-ai	120	2	d	d	LS
07-kim-ai	120	3	)	)	-RRB-
07-kim-ai	120	4	Opacity	opacity	NN
07-kim-ai	120	5	in	in	IN
07-kim-ai	120	6	decision	decision	NN
07-kim-ai	120	7	-	-	HYPH
07-kim-ai	120	8	making	make	VBG
07-kim-ai	120	9	Machine	Machine	NNP
07-kim-ai	120	10	learning	learning	NN
07-kim-ai	120	11	is	be	VBZ
07-kim-ai	120	12	used	use	VBN
07-kim-ai	120	13	to	to	TO
07-kim-ai	120	14	build	build	VB
07-kim-ai	120	15	many	many	JJ
07-kim-ai	120	16	AI	AI	NNP
07-kim-ai	120	17	systems	system	NNS
07-kim-ai	120	18	today	today	NN
07-kim-ai	120	19	.	.	.
07-kim-ai	121	1	Instead	instead	RB
07-kim-ai	121	2	of	of	IN
07-kim-ai	121	3	prescribing	prescribe	VBG
07-kim-ai	121	4	a	a	DT
07-kim-ai	121	5	pre	pre	JJ
07-kim-ai	121	6	-	-	JJ
07-kim-ai	121	7	determined	determined	JJ
07-kim-ai	121	8	algorithm	algorithm	NN
07-kim-ai	121	9	,	,	,
07-kim-ai	121	10	a	a	DT
07-kim-ai	121	11	machine	machine	NN
07-kim-ai	121	12	learning	learning	NN
07-kim-ai	121	13	system	system	NN
07-kim-ai	121	14	goes	go	VBZ
07-kim-ai	121	15	through	through	IN
07-kim-ai	121	16	a	a	DT
07-kim-ai	121	17	so	so	RB
07-kim-ai	121	18	-	-	HYPH
07-kim-ai	121	19	called	call	VBN
07-kim-ai	121	20	‘	'	``
07-kim-ai	121	21	training	training	NN
07-kim-ai	121	22	’	'	''
07-kim-ai	121	23	process	process	NN
07-kim-ai	121	24	to	to	TO
07-kim-ai	121	25	produce	produce	VB
07-kim-ai	121	26	the	the	DT
07-kim-ai	121	27	final	final	JJ
07-kim-ai	121	28	algorithm	algorithm	NN
07-kim-ai	121	29	from	from	IN
07-kim-ai	121	30	a	a	DT
07-kim-ai	121	31	large	large	JJ
07-kim-ai	121	32	amount	amount	NN
07-kim-ai	121	33	of	of	IN
07-kim-ai	121	34	data	datum	NNS
07-kim-ai	121	35	.	.	.
07-kim-ai	122	1	For	for	IN
07-kim-ai	122	2	example	example	NN
07-kim-ai	122	3	,	,	,
07-kim-ai	122	4	a	a	DT
07-kim-ai	122	5	machine	machine	NN
07-kim-ai	122	6	learning	learning	NN
07-kim-ai	122	7	system	system	NN
07-kim-ai	122	8	may	may	MD
07-kim-ai	122	9	generate	generate	VB
07-kim-ai	122	10	an	an	DT
07-kim-ai	122	11	algorithm	algorithm	NN
07-kim-ai	122	12	that	that	WDT
07-kim-ai	122	13	successfully	successfully	RB
07-kim-ai	122	14	recognizes	recognize	VBZ
07-kim-ai	122	15	cats	cat	NNS
07-kim-ai	122	16	in	in	IN
07-kim-ai	122	17	a	a	DT
07-kim-ai	122	18	photo	photo	NN
07-kim-ai	122	19	after	after	IN
07-kim-ai	122	20	going	go	VBG
07-kim-ai	122	21	through	through	IN
07-kim-ai	122	22	millions	million	NNS
07-kim-ai	122	23	of	of	IN
07-kim-ai	122	24	photos	photo	NNS
07-kim-ai	122	25	that	that	WDT
07-kim-ai	122	26	show	show	VBP
07-kim-ai	122	27	cats	cat	NNS
07-kim-ai	122	28	in	in	IN
07-kim-ai	122	29	many	many	JJ
07-kim-ai	122	30	different	different	JJ
07-kim-ai	122	31	postures	posture	NNS
07-kim-ai	122	32	from	from	IN
07-kim-ai	122	33	various	various	JJ
07-kim-ai	122	34	angles.14	angles.14	NNP
07-kim-ai	122	35	But	but	CC
07-kim-ai	122	36	the	the	DT
07-kim-ai	122	37	resulting	result	VBG
07-kim-ai	122	38	algorithm	algorithm	NN
07-kim-ai	122	39	is	be	VBZ
07-kim-ai	122	40	a	a	DT
07-kim-ai	122	41	complex	complex	JJ
07-kim-ai	122	42	mathematical	mathematical	JJ
07-kim-ai	122	43	formula	formula	NN
07-kim-ai	122	44	and	and	CC
07-kim-ai	122	45	not	not	RB
07-kim-ai	122	46	something	something	NN
07-kim-ai	122	47	that	that	WDT
07-kim-ai	122	48	humans	human	NNS
07-kim-ai	122	49	can	can	MD
07-kim-ai	122	50	easily	easily	RB
07-kim-ai	122	51	decipher	decipher	VB
07-kim-ai	122	52	.	.	.
07-kim-ai	123	1	This	this	DT
07-kim-ai	123	2	means	mean	VBZ
07-kim-ai	123	3	that	that	IN
07-kim-ai	123	4	the	the	DT
07-kim-ai	123	5	inner	inner	JJ
07-kim-ai	123	6	workings	working	NNS
07-kim-ai	123	7	of	of	IN
07-kim-ai	123	8	a	a	DT
07-kim-ai	123	9	machine	machine	NN
07-kim-ai	123	10	learning	learn	VBG
07-kim-ai	123	11	AI	ai	NN
07-kim-ai	123	12	system	system	NN
07-kim-ai	123	13	and	and	CC
07-kim-ai	123	14	its	-PRON-	PRP$
07-kim-ai	123	15	decision	decision	NN
07-kim-ai	123	16	-	-	HYPH
07-kim-ai	123	17	making	make	VBG
07-kim-ai	123	18	process	process	NN
07-kim-ai	123	19	is	be	VBZ
07-kim-ai	123	20	opaque	opaque	JJ
07-kim-ai	123	21	to	to	IN
07-kim-ai	123	22	human	human	JJ
07-kim-ai	123	23	understanding	understanding	NN
07-kim-ai	123	24	,	,	,
07-kim-ai	123	25	even	even	RB
07-kim-ai	123	26	to	to	IN
07-kim-ai	123	27	those	those	DT
07-kim-ai	123	28	who	who	WP
07-kim-ai	123	29	built	build	VBD
07-kim-ai	123	30	the	the	DT
07-kim-ai	123	31	system	system	NN
07-kim-ai	123	32	itself	-PRON-	PRP
07-kim-ai	123	33	(	(	-LRB-
07-kim-ai	123	34	Knight	Knight	NNP
07-kim-ai	123	35	2017	2017	CD
07-kim-ai	123	36	)	)	-RRB-
07-kim-ai	123	37	.	.	.
07-kim-ai	124	1	In	in	IN
07-kim-ai	124	2	cases	case	NNS
07-kim-ai	124	3	where	where	WRB
07-kim-ai	124	4	the	the	DT
07-kim-ai	124	5	actions	action	NNS
07-kim-ai	124	6	of	of	IN
07-kim-ai	124	7	an	an	DT
07-kim-ai	124	8	AI	AI	NNP
07-kim-ai	124	9	system	system	NN
07-kim-ai	124	10	can	can	MD
07-kim-ai	124	11	have	have	VB
07-kim-ai	124	12	grave	grave	JJ
07-kim-ai	124	13	consequences	consequence	NNS
07-kim-ai	124	14	such	such	JJ
07-kim-ai	124	15	as	as	IN
07-kim-ai	124	16	a	a	DT
07-kim-ai	124	17	military	military	JJ
07-kim-ai	124	18	robot	robot	NN
07-kim-ai	124	19	,	,	,
07-kim-ai	124	20	such	such	JJ
07-kim-ai	124	21	opacity	opacity	NN
07-kim-ai	124	22	becomes	become	VBZ
07-kim-ai	124	23	a	a	DT
07-kim-ai	124	24	serious	serious	JJ
07-kim-ai	124	25	problem.15	problem.15	NNP
07-kim-ai	124	26	13(Kahn	13(Kahn	NNP
07-kim-ai	124	27	2012	2012	CD
07-kim-ai	124	28	)	)	-RRB-
07-kim-ai	124	29	also	also	RB
07-kim-ai	124	30	argues	argue	VBZ
07-kim-ai	124	31	that	that	IN
07-kim-ai	124	32	the	the	DT
07-kim-ai	124	33	resulting	result	VBG
07-kim-ai	124	34	increase	increase	NN
07-kim-ai	124	35	in	in	IN
07-kim-ai	124	36	the	the	DT
07-kim-ai	124	37	number	number	NN
07-kim-ai	124	38	of	of	IN
07-kim-ai	124	39	wars	war	NNS
07-kim-ai	124	40	by	by	IN
07-kim-ai	124	41	the	the	DT
07-kim-ai	124	42	use	use	NN
07-kim-ai	124	43	of	of	IN
07-kim-ai	124	44	military	military	JJ
07-kim-ai	124	45	robots	robot	NNS
07-kim-ai	124	46	will	will	MD
07-kim-ai	124	47	be	be	VB
07-kim-ai	124	48	morally	morally	RB
07-kim-ai	124	49	bad	bad	JJ
07-kim-ai	124	50	.	.	.
07-kim-ai	125	1	14Google	14google	CD
07-kim-ai	125	2	’s	’s	POS
07-kim-ai	125	3	research	research	NN
07-kim-ai	125	4	team	team	NN
07-kim-ai	125	5	created	create	VBD
07-kim-ai	125	6	an	an	DT
07-kim-ai	125	7	AI	AI	NNP
07-kim-ai	125	8	algorithm	algorithm	NN
07-kim-ai	125	9	that	that	WDT
07-kim-ai	125	10	learned	learn	VBD
07-kim-ai	125	11	how	how	WRB
07-kim-ai	125	12	to	to	TO
07-kim-ai	125	13	recognize	recognize	VB
07-kim-ai	125	14	a	a	DT
07-kim-ai	125	15	cat	cat	NN
07-kim-ai	125	16	in	in	IN
07-kim-ai	125	17	2012	2012	CD
07-kim-ai	125	18	.	.	.
07-kim-ai	126	1	The	the	DT
07-kim-ai	126	2	neural	neural	JJ
07-kim-ai	126	3	network	network	NN
07-kim-ai	126	4	behind	behind	IN
07-kim-ai	126	5	this	this	DT
07-kim-ai	126	6	algorithm	algorithm	NN
07-kim-ai	126	7	had	have	VBD
07-kim-ai	126	8	an	an	DT
07-kim-ai	126	9	array	array	NN
07-kim-ai	126	10	of	of	IN
07-kim-ai	126	11	16,000	16,000	CD
07-kim-ai	126	12	processors	processor	NNS
07-kim-ai	126	13	and	and	CC
07-kim-ai	126	14	more	more	JJR
07-kim-ai	126	15	than	than	IN
07-kim-ai	126	16	one	one	CD
07-kim-ai	126	17	billion	billion	CD
07-kim-ai	126	18	connections	connection	NNS
07-kim-ai	126	19	.	.	.
07-kim-ai	127	1	Unlabeled	unlabeled	JJ
07-kim-ai	127	2	random	random	JJ
07-kim-ai	127	3	thumbnail	thumbnail	NN
07-kim-ai	127	4	images	image	NNS
07-kim-ai	127	5	from	from	IN
07-kim-ai	127	6	10	10	CD
07-kim-ai	127	7	million	million	CD
07-kim-ai	127	8	YouTube	YouTube	NNP
07-kim-ai	127	9	videos	video	NNS
07-kim-ai	127	10	allowed	allow	VBD
07-kim-ai	127	11	this	this	DT
07-kim-ai	127	12	algorithm	algorithm	NN
07-kim-ai	127	13	to	to	TO
07-kim-ai	127	14	learn	learn	VB
07-kim-ai	127	15	to	to	TO
07-kim-ai	127	16	identify	identify	VB
07-kim-ai	127	17	cats	cat	NNS
07-kim-ai	127	18	by	by	IN
07-kim-ai	127	19	itself	-PRON-	PRP
07-kim-ai	127	20	.	.	.
07-kim-ai	128	1	See	see	VB
07-kim-ai	128	2	Markoff	Markoff	NNP
07-kim-ai	128	3	2012	2012	CD
07-kim-ai	128	4	and	and	CC
07-kim-ai	128	5	Clark	Clark	NNP
07-kim-ai	128	6	2012	2012	CD
07-kim-ai	128	7	.	.	.
07-kim-ai	129	1	15This	15this	CD
07-kim-ai	129	2	black	black	JJ
07-kim-ai	129	3	-	-	HYPH
07-kim-ai	129	4	box	box	NN
07-kim-ai	129	5	nature	nature	NN
07-kim-ai	129	6	of	of	IN
07-kim-ai	129	7	AI	AI	NNP
07-kim-ai	129	8	systems	system	NNS
07-kim-ai	129	9	powered	power	VBN
07-kim-ai	129	10	by	by	IN
07-kim-ai	129	11	machine	machine	NN
07-kim-ai	129	12	learning	learning	NN
07-kim-ai	129	13	has	have	VBZ
07-kim-ai	129	14	raised	raise	VBN
07-kim-ai	129	15	great	great	JJ
07-kim-ai	129	16	concern	concern	NN
07-kim-ai	129	17	among	among	IN
07-kim-ai	129	18	many	many	JJ
07-kim-ai	129	19	AI	AI	NNP
07-kim-ai	129	20	re-	re-	JJ
07-kim-ai	129	21	searchers	searcher	NNS
07-kim-ai	129	22	in	in	IN
07-kim-ai	129	23	recent	recent	JJ
07-kim-ai	129	24	years	year	NNS
07-kim-ai	129	25	.	.	.
07-kim-ai	130	1	This	this	DT
07-kim-ai	130	2	is	be	VBZ
07-kim-ai	130	3	problematic	problematic	JJ
07-kim-ai	130	4	in	in	IN
07-kim-ai	130	5	all	all	DT
07-kim-ai	130	6	areas	area	NNS
07-kim-ai	130	7	where	where	WRB
07-kim-ai	130	8	these	these	DT
07-kim-ai	130	9	AI	AI	NNP
07-kim-ai	130	10	systems	system	NNS
07-kim-ai	130	11	are	be	VBP
07-kim-ai	130	12	used	use	VBN
07-kim-ai	130	13	for	for	IN
07-kim-ai	130	14	decision	decision	NN
07-kim-ai	130	15	-	-	HYPH
07-kim-ai	130	16	making	making	NN
07-kim-ai	130	17	,	,	,
07-kim-ai	130	18	not	not	RB
07-kim-ai	130	19	just	just	RB
07-kim-ai	130	20	in	in	IN
07-kim-ai	130	21	military	military	JJ
07-kim-ai	130	22	operations	operation	NNS
07-kim-ai	130	23	.	.	.
07-kim-ai	131	1	The	the	DT
07-kim-ai	131	2	gravity	gravity	NN
07-kim-ai	131	3	of	of	IN
07-kim-ai	131	4	decisions	decision	NNS
07-kim-ai	131	5	made	make	VBN
07-kim-ai	131	6	in	in	IN
07-kim-ai	131	7	a	a	DT
07-kim-ai	131	8	military	military	JJ
07-kim-ai	131	9	operation	operation	NN
07-kim-ai	131	10	makes	make	VBZ
07-kim-ai	131	11	this	this	DT
07-kim-ai	131	12	problem	problem	NN
07-kim-ai	131	13	even	even	RB
07-kim-ai	131	14	more	more	RBR
07-kim-ai	131	15	troublesome	troublesome	JJ
07-kim-ai	131	16	.	.	.
07-kim-ai	132	1	Fortunately	fortunately	RB
07-kim-ai	132	2	,	,	,
07-kim-ai	132	3	some	some	DT
07-kim-ai	132	4	AI	AI	NNP
07-kim-ai	132	5	researchers	researcher	NNS
07-kim-ai	132	6	including	include	VBG
07-kim-ai	132	7	those	those	DT
07-kim-ai	132	8	in	in	IN
07-kim-ai	132	9	the	the	DT
07-kim-ai	132	10	US	US	NNP
07-kim-ai	132	11	Department	Department	NNP
07-kim-ai	132	12	of	of	IN
07-kim-ai	132	13	Defense	Defense	NNP
07-kim-ai	132	14	are	be	VBP
07-kim-ai	132	15	actively	actively	RB
07-kim-ai	132	16	working	work	VBG
07-kim-ai	132	17	to	to	TO
07-kim-ai	132	18	make	make	VB
07-kim-ai	132	19	AI	AI	NNP
07-kim-ai	132	20	sys-	sys-	RB
07-kim-ai	132	21	tems	tem	NNS
07-kim-ai	132	22	explainable	explainable	JJ
07-kim-ai	132	23	.	.	.
07-kim-ai	133	1	But	but	CC
07-kim-ai	133	2	until	until	IN
07-kim-ai	133	3	such	such	JJ
07-kim-ai	133	4	research	research	NN
07-kim-ai	133	5	bears	bear	VBZ
07-kim-ai	133	6	fruit	fruit	NN
07-kim-ai	133	7	and	and	CC
07-kim-ai	133	8	AI	AI	NNP
07-kim-ai	133	9	systems	system	NNS
07-kim-ai	133	10	become	become	VBP
07-kim-ai	133	11	fully	fully	RB
07-kim-ai	133	12	explainable	explainable	JJ
07-kim-ai	133	13	,	,	,
07-kim-ai	133	14	their	-PRON-	PRP$
07-kim-ai	133	15	military	military	JJ
07-kim-ai	133	16	use	use	NN
07-kim-ai	133	17	means	mean	VBZ
07-kim-ai	133	18	accepting	accept	VBG
07-kim-ai	133	19	many	many	JJ
07-kim-ai	133	20	unknown	unknown	JJ
07-kim-ai	133	21	variables	variable	NNS
07-kim-ai	133	22	and	and	CC
07-kim-ai	133	23	unforeseeable	unforeseeable	JJ
07-kim-ai	133	24	consequences	consequence	NNS
07-kim-ai	133	25	.	.	.
07-kim-ai	134	1	See	see	VB
07-kim-ai	134	2	Turek	Turek	NNP
07-kim-ai	134	3	n.d	n.d	NNP
07-kim-ai	134	4	.	.	.
07-kim-ai	134	5	78	78	CD
07-kim-ai	134	6	Machine	Machine	NNP
07-kim-ai	134	7	Learning	Learning	NNP
07-kim-ai	134	8	,	,	,
07-kim-ai	134	9	Libraries	Libraries	NNPS
07-kim-ai	134	10	,	,	,
07-kim-ai	134	11	and	and	CC
07-kim-ai	134	12	Cross	Cross	NNP
07-kim-ai	134	13	-	-	NNP
07-kim-ai	134	14	Disciplinary	Disciplinary	NNP
07-kim-ai	134	15	ResearchǔChapter	ResearchǔChapter	NNP
07-kim-ai	134	16	7	7	CD
07-kim-ai	134	17	AI	AI	NNP
07-kim-ai	134	18	Applications	Applications	NNPS
07-kim-ai	134	19	for	for	IN
07-kim-ai	134	20	Libraries	library	NNS
07-kim-ai	134	21	Do	do	VBP
07-kim-ai	134	22	these	these	DT
07-kim-ai	134	23	ethical	ethical	JJ
07-kim-ai	134	24	concerns	concern	NNS
07-kim-ai	134	25	outlined	outline	VBN
07-kim-ai	134	26	above	above	IN
07-kim-ai	134	27	apply	apply	VBP
07-kim-ai	134	28	to	to	IN
07-kim-ai	134	29	libraries	library	NNS
07-kim-ai	134	30	?	?	.
07-kim-ai	135	1	To	to	TO
07-kim-ai	135	2	answer	answer	VB
07-kim-ai	135	3	that	that	DT
07-kim-ai	135	4	,	,	,
07-kim-ai	135	5	let	let	VB
07-kim-ai	135	6	us	-PRON-	PRP
07-kim-ai	135	7	first	first	RB
07-kim-ai	135	8	take	take	VB
07-kim-ai	135	9	a	a	DT
07-kim-ai	135	10	look	look	NN
07-kim-ai	135	11	at	at	IN
07-kim-ai	135	12	how	how	WRB
07-kim-ai	135	13	AI	AI	NNP
07-kim-ai	135	14	,	,	,
07-kim-ai	135	15	particularly	particularly	RB
07-kim-ai	135	16	machine	machine	NN
07-kim-ai	135	17	learning	learning	NN
07-kim-ai	135	18	,	,	,
07-kim-ai	135	19	may	may	MD
07-kim-ai	135	20	apply	apply	VB
07-kim-ai	135	21	to	to	IN
07-kim-ai	135	22	library	library	NN
07-kim-ai	135	23	services	service	NNS
07-kim-ai	135	24	and	and	CC
07-kim-ai	135	25	operations	operation	NNS
07-kim-ai	135	26	.	.	.
07-kim-ai	136	1	AI-	AI-	NNP
07-kim-ai	136	2	powered	power	VBN
07-kim-ai	136	3	digital	digital	JJ
07-kim-ai	136	4	assistants	assistant	NNS
07-kim-ai	136	5	are	be	VBP
07-kim-ai	136	6	likely	likely	JJ
07-kim-ai	136	7	to	to	TO
07-kim-ai	136	8	mediate	mediate	VB
07-kim-ai	136	9	a	a	DT
07-kim-ai	136	10	library	library	NN
07-kim-ai	136	11	user	user	NN
07-kim-ai	136	12	’s	’s	POS
07-kim-ai	136	13	information	information	NN
07-kim-ai	136	14	search	search	NN
07-kim-ai	136	15	,	,	,
07-kim-ai	136	16	discovery	discovery	NN
07-kim-ai	136	17	,	,	,
07-kim-ai	136	18	and	and	CC
07-kim-ai	136	19	retrieval	retrieval	NN
07-kim-ai	136	20	activities	activity	NNS
07-kim-ai	136	21	in	in	IN
07-kim-ai	136	22	the	the	DT
07-kim-ai	136	23	near	near	JJ
07-kim-ai	136	24	future	future	NN
07-kim-ai	136	25	.	.	.
07-kim-ai	137	1	In	in	IN
07-kim-ai	137	2	recent	recent	JJ
07-kim-ai	137	3	years	year	NNS
07-kim-ai	137	4	,	,	,
07-kim-ai	137	5	machine	machine	NN
07-kim-ai	137	6	learning	learning	NN
07-kim-ai	137	7	and	and	CC
07-kim-ai	137	8	deep	deep	JJ
07-kim-ai	137	9	learning	learning	NN
07-kim-ai	137	10	have	have	VBP
07-kim-ai	137	11	brought	bring	VBN
07-kim-ai	137	12	significant	significant	JJ
07-kim-ai	137	13	improvement	improvement	NN
07-kim-ai	137	14	to	to	IN
07-kim-ai	137	15	natural	natural	JJ
07-kim-ai	137	16	language	language	NN
07-kim-ai	137	17	processing	processing	NN
07-kim-ai	137	18	(	(	-LRB-
07-kim-ai	137	19	NLP	NLP	NNP
07-kim-ai	137	20	)	)	-RRB-
07-kim-ai	137	21	,	,	,
07-kim-ai	137	22	which	which	WDT
07-kim-ai	137	23	deals	deal	VBZ
07-kim-ai	137	24	with	with	IN
07-kim-ai	137	25	analyzing	analyze	VBG
07-kim-ai	137	26	large	large	JJ
07-kim-ai	137	27	amounts	amount	NNS
07-kim-ai	137	28	of	of	IN
07-kim-ai	137	29	natural	natural	JJ
07-kim-ai	137	30	lan-	lan-	NN
07-kim-ai	137	31	guage	guage	NN
07-kim-ai	137	32	data	datum	NNS
07-kim-ai	137	33	to	to	TO
07-kim-ai	137	34	make	make	VB
07-kim-ai	137	35	the	the	DT
07-kim-ai	137	36	interaction	interaction	NN
07-kim-ai	137	37	between	between	IN
07-kim-ai	137	38	people	people	NNS
07-kim-ai	137	39	and	and	CC
07-kim-ai	137	40	machines	machine	NNS
07-kim-ai	137	41	in	in	IN
07-kim-ai	137	42	natural	natural	JJ
07-kim-ai	137	43	languages	language	NNS
07-kim-ai	137	44	possible	possible	JJ
07-kim-ai	137	45	.	.	.
07-kim-ai	138	1	For	for	IN
07-kim-ai	138	2	instance	instance	NN
07-kim-ai	138	3	,	,	,
07-kim-ai	138	4	Google	Google	NNP
07-kim-ai	138	5	Assistant	Assistant	NNP
07-kim-ai	138	6	’s	’s	POS
07-kim-ai	138	7	new	new	JJ
07-kim-ai	138	8	feature	feature	NN
07-kim-ai	138	9	‘	'	``
07-kim-ai	138	10	duplex	duplex	NN
07-kim-ai	138	11	’	'	''
07-kim-ai	138	12	was	be	VBD
07-kim-ai	138	13	shown	show	VBN
07-kim-ai	138	14	to	to	TO
07-kim-ai	138	15	successfully	successfully	RB
07-kim-ai	138	16	make	make	VB
07-kim-ai	138	17	a	a	DT
07-kim-ai	138	18	phone	phone	NN
07-kim-ai	138	19	reservation	reservation	NN
07-kim-ai	138	20	with	with	IN
07-kim-ai	138	21	restaurant	restaurant	NN
07-kim-ai	138	22	staff	staff	NN
07-kim-ai	138	23	in	in	IN
07-kim-ai	138	24	2018	2018	CD
07-kim-ai	138	25	(	(	-LRB-
07-kim-ai	138	26	Welch	Welch	NNP
07-kim-ai	138	27	2018	2018	CD
07-kim-ai	138	28	)	)	-RRB-
07-kim-ai	138	29	.	.	.
07-kim-ai	139	1	Google	Google	NNP
07-kim-ai	139	2	’s	’s	POS
07-kim-ai	139	3	real	real	JJ
07-kim-ai	139	4	-	-	HYPH
07-kim-ai	139	5	time	time	NN
07-kim-ai	139	6	translation	translation	NN
07-kim-ai	139	7	capability	capability	NN
07-kim-ai	139	8	for	for	IN
07-kim-ai	139	9	44	44	CD
07-kim-ai	139	10	different	different	JJ
07-kim-ai	139	11	languages	language	NNS
07-kim-ai	139	12	was	be	VBD
07-kim-ai	139	13	introduced	introduce	VBN
07-kim-ai	139	14	to	to	IN
07-kim-ai	139	15	Google	Google	NNP
07-kim-ai	139	16	Assist	Assist	NNP
07-kim-ai	139	17	-	-	HYPH
07-kim-ai	139	18	enabled	enable	VBD
07-kim-ai	139	19	Android	Android	NNP
07-kim-ai	139	20	and	and	CC
07-kim-ai	139	21	iOS	iOS	NNP
07-kim-ai	139	22	phones	phone	NNS
07-kim-ai	139	23	in	in	IN
07-kim-ai	139	24	2019	2019	CD
07-kim-ai	139	25	(	(	-LRB-
07-kim-ai	139	26	Rincon	Rincon	NNP
07-kim-ai	139	27	2019	2019	CD
07-kim-ai	139	28	)	)	-RRB-
07-kim-ai	139	29	.	.	.
07-kim-ai	140	1	As	as	IN
07-kim-ai	140	2	digital	digital	JJ
07-kim-ai	140	3	assistants	assistant	NNS
07-kim-ai	140	4	become	become	VBP
07-kim-ai	140	5	capable	capable	JJ
07-kim-ai	140	6	of	of	IN
07-kim-ai	140	7	handling	handle	VBG
07-kim-ai	140	8	more	more	RBR
07-kim-ai	140	9	sophisticated	sophisticated	JJ
07-kim-ai	140	10	language	language	NN
07-kim-ai	140	11	tasks	task	NNS
07-kim-ai	140	12	,	,	,
07-kim-ai	140	13	their	-PRON-	PRP$
07-kim-ai	140	14	use	use	NN
07-kim-ai	140	15	as	as	IN
07-kim-ai	140	16	a	a	DT
07-kim-ai	140	17	flexible	flexible	JJ
07-kim-ai	140	18	voice	voice	NN
07-kim-ai	140	19	user	user	NN
07-kim-ai	140	20	interface	interface	NN
07-kim-ai	140	21	will	will	MD
07-kim-ai	140	22	only	only	RB
07-kim-ai	140	23	increase	increase	VB
07-kim-ai	140	24	.	.	.
07-kim-ai	141	1	Such	such	JJ
07-kim-ai	141	2	digital	digital	JJ
07-kim-ai	141	3	assistants	assistant	NNS
07-kim-ai	141	4	will	will	MD
07-kim-ai	141	5	be	be	VB
07-kim-ai	141	6	able	able	JJ
07-kim-ai	141	7	to	to	TO
07-kim-ai	141	8	directly	directly	RB
07-kim-ai	141	9	interact	interact	VB
07-kim-ai	141	10	with	with	IN
07-kim-ai	141	11	library	library	JJ
07-kim-ai	141	12	systems	system	NNS
07-kim-ai	141	13	and	and	CC
07-kim-ai	141	14	applications	application	NNS
07-kim-ai	141	15	,	,	,
07-kim-ai	141	16	automatically	automatically	RB
07-kim-ai	141	17	interpret	interpret	VB
07-kim-ai	141	18	a	a	DT
07-kim-ai	141	19	query	query	NN
07-kim-ai	141	20	,	,	,
07-kim-ai	141	21	and	and	CC
07-kim-ai	141	22	return	return	VB
07-kim-ai	141	23	results	result	NNS
07-kim-ai	141	24	that	that	WDT
07-kim-ai	141	25	they	-PRON-	PRP
07-kim-ai	141	26	deem	deem	VBP
07-kim-ai	141	27	to	to	TO
07-kim-ai	141	28	be	be	VB
07-kim-ai	141	29	most	most	RBS
07-kim-ai	141	30	relevant	relevant	JJ
07-kim-ai	141	31	.	.	.
07-kim-ai	142	1	Those	those	DT
07-kim-ai	142	2	digital	digital	JJ
07-kim-ai	142	3	assistants	assistant	NNS
07-kim-ai	142	4	can	can	MD
07-kim-ai	142	5	also	also	RB
07-kim-ai	142	6	be	be	VB
07-kim-ai	142	7	equipped	equip	VBN
07-kim-ai	142	8	to	to	TO
07-kim-ai	142	9	handle	handle	VB
07-kim-ai	142	10	the	the	DT
07-kim-ai	142	11	library	library	NN
07-kim-ai	142	12	’s	’s	,
07-kim-ai	142	13	traditional	traditional	JJ
07-kim-ai	142	14	reference	reference	NN
07-kim-ai	142	15	or	or	CC
07-kim-ai	142	16	readers	reader	NNS
07-kim-ai	142	17	’	’	POS
07-kim-ai	142	18	advisory	advisory	JJ
07-kim-ai	142	19	service	service	NN
07-kim-ai	142	20	.	.	.
07-kim-ai	143	1	Integrated	integrate	VBN
07-kim-ai	143	2	into	into	IN
07-kim-ai	143	3	a	a	DT
07-kim-ai	143	4	humanoid	humanoid	JJ
07-kim-ai	143	5	robot	robot	NN
07-kim-ai	143	6	body	body	NN
07-kim-ai	143	7	,	,	,
07-kim-ai	143	8	they	-PRON-	PRP
07-kim-ai	143	9	may	may	MD
07-kim-ai	143	10	even	even	RB
07-kim-ai	143	11	greet	greet	VB
07-kim-ai	143	12	library	library	JJ
07-kim-ai	143	13	patrons	patron	NNS
07-kim-ai	143	14	at	at	IN
07-kim-ai	143	15	the	the	DT
07-kim-ai	143	16	entrance	entrance	NN
07-kim-ai	143	17	and	and	CC
07-kim-ai	143	18	answer	answer	VB
07-kim-ai	143	19	directional	directional	JJ
07-kim-ai	143	20	questions	question	NNS
07-kim-ai	143	21	about	about	IN
07-kim-ai	143	22	the	the	DT
07-kim-ai	143	23	library	library	NN
07-kim-ai	143	24	building	building	NN
07-kim-ai	143	25	.	.	.
07-kim-ai	144	1	Cataloging	catalog	VBG
07-kim-ai	144	2	,	,	,
07-kim-ai	144	3	abstracting	abstracting	NN
07-kim-ai	144	4	,	,	,
07-kim-ai	144	5	and	and	CC
07-kim-ai	144	6	indexing	indexing	NN
07-kim-ai	144	7	are	be	VBP
07-kim-ai	144	8	other	other	JJ
07-kim-ai	144	9	areas	area	NNS
07-kim-ai	144	10	where	where	WRB
07-kim-ai	144	11	AI	AI	NNP
07-kim-ai	144	12	will	will	MD
07-kim-ai	144	13	be	be	VB
07-kim-ai	144	14	actively	actively	RB
07-kim-ai	144	15	utilized	utilize	VBN
07-kim-ai	144	16	.	.	.
07-kim-ai	145	1	Cur-	cur-	VB
07-kim-ai	145	2	rently	rently	RB
07-kim-ai	145	3	,	,	,
07-kim-ai	145	4	those	those	DT
07-kim-ai	145	5	tasks	task	NNS
07-kim-ai	145	6	are	be	VBP
07-kim-ai	145	7	performed	perform	VBN
07-kim-ai	145	8	by	by	IN
07-kim-ai	145	9	skilled	skilled	JJ
07-kim-ai	145	10	professionals	professional	NNS
07-kim-ai	145	11	.	.	.
07-kim-ai	146	1	But	but	CC
07-kim-ai	146	2	as	as	IN
07-kim-ai	146	3	AI	AI	NNP
07-kim-ai	146	4	applications	application	NNS
07-kim-ai	146	5	become	become	VBP
07-kim-ai	146	6	more	more	RBR
07-kim-ai	146	7	sophisticated	sophisticated	JJ
07-kim-ai	146	8	,	,	,
07-kim-ai	146	9	we	-PRON-	PRP
07-kim-ai	146	10	may	may	MD
07-kim-ai	146	11	see	see	VB
07-kim-ai	146	12	many	many	JJ
07-kim-ai	146	13	of	of	IN
07-kim-ai	146	14	those	those	DT
07-kim-ai	146	15	tasks	task	NNS
07-kim-ai	146	16	partially	partially	RB
07-kim-ai	146	17	or	or	CC
07-kim-ai	146	18	fully	fully	RB
07-kim-ai	146	19	automated	automate	VBN
07-kim-ai	146	20	and	and	CC
07-kim-ai	146	21	handed	hand	VBD
07-kim-ai	146	22	over	over	RP
07-kim-ai	146	23	to	to	IN
07-kim-ai	146	24	AI	AI	NNP
07-kim-ai	146	25	systems	system	NNS
07-kim-ai	146	26	.	.	.
07-kim-ai	147	1	Machine	machine	NN
07-kim-ai	147	2	learning	learning	NN
07-kim-ai	147	3	and	and	CC
07-kim-ai	147	4	deep	deep	JJ
07-kim-ai	147	5	learning	learning	NN
07-kim-ai	147	6	can	can	MD
07-kim-ai	147	7	be	be	VB
07-kim-ai	147	8	used	use	VBN
07-kim-ai	147	9	to	to	TO
07-kim-ai	147	10	extract	extract	VB
07-kim-ai	147	11	key	key	JJ
07-kim-ai	147	12	information	information	NN
07-kim-ai	147	13	from	from	IN
07-kim-ai	147	14	a	a	DT
07-kim-ai	147	15	large	large	JJ
07-kim-ai	147	16	number	number	NN
07-kim-ai	147	17	of	of	IN
07-kim-ai	147	18	documents	document	NNS
07-kim-ai	147	19	or	or	CC
07-kim-ai	147	20	from	from	IN
07-kim-ai	147	21	information	information	NN
07-kim-ai	147	22	-	-	HYPH
07-kim-ai	147	23	rich	rich	JJ
07-kim-ai	147	24	visual	visual	JJ
07-kim-ai	147	25	materials	material	NNS
07-kim-ai	147	26	,	,	,
07-kim-ai	147	27	such	such	JJ
07-kim-ai	147	28	as	as	IN
07-kim-ai	147	29	maps	map	NNS
07-kim-ai	147	30	and	and	CC
07-kim-ai	147	31	video	video	NN
07-kim-ai	147	32	recordings	recording	NNS
07-kim-ai	147	33	,	,	,
07-kim-ai	147	34	and	and	CC
07-kim-ai	147	35	generate	generate	VB
07-kim-ai	147	36	metadata	metadata	NN
07-kim-ai	147	37	or	or	CC
07-kim-ai	147	38	a	a	DT
07-kim-ai	147	39	summary	summary	NN
07-kim-ai	147	40	.	.	.
07-kim-ai	148	1	Since	since	IN
07-kim-ai	148	2	machine	machine	NN
07-kim-ai	148	3	learning	learning	NN
07-kim-ai	148	4	is	be	VBZ
07-kim-ai	148	5	new	new	JJ
07-kim-ai	148	6	to	to	IN
07-kim-ai	148	7	libraries	library	NNS
07-kim-ai	148	8	,	,	,
07-kim-ai	148	9	there	there	EX
07-kim-ai	148	10	are	be	VBP
07-kim-ai	148	11	a	a	DT
07-kim-ai	148	12	relatively	relatively	RB
07-kim-ai	148	13	small	small	JJ
07-kim-ai	148	14	number	number	NN
07-kim-ai	148	15	of	of	IN
07-kim-ai	148	16	machine	machine	NN
07-kim-ai	148	17	learning	learning	NN
07-kim-ai	148	18	applications	application	NNS
07-kim-ai	148	19	developed	develop	VBN
07-kim-ai	148	20	for	for	IN
07-kim-ai	148	21	libraries	library	NNS
07-kim-ai	148	22	’	’	POS
07-kim-ai	148	23	use	use	NN
07-kim-ai	148	24	.	.	.
07-kim-ai	149	1	They	-PRON-	PRP
07-kim-ai	149	2	are	be	VBP
07-kim-ai	149	3	likely	likely	JJ
07-kim-ai	149	4	to	to	TO
07-kim-ai	149	5	grow	grow	VB
07-kim-ai	149	6	in	in	IN
07-kim-ai	149	7	number	number	NN
07-kim-ai	149	8	.	.	.
07-kim-ai	150	1	Yewno	Yewno	NNP
07-kim-ai	150	2	,	,	,
07-kim-ai	150	3	Quartolio	Quartolio	NNP
07-kim-ai	150	4	,	,	,
07-kim-ai	150	5	and	and	CC
07-kim-ai	150	6	Iris.ai	iris.ai	NN
07-kim-ai	150	7	are	be	VBP
07-kim-ai	150	8	examples	example	NNS
07-kim-ai	150	9	of	of	IN
07-kim-ai	150	10	the	the	DT
07-kim-ai	150	11	commercial	commercial	JJ
07-kim-ai	150	12	products	product	NNS
07-kim-ai	150	13	developed	develop	VBN
07-kim-ai	150	14	with	with	IN
07-kim-ai	150	15	machine	machine	NN
07-kim-ai	150	16	learning	learning	NN
07-kim-ai	150	17	and	and	CC
07-kim-ai	150	18	deep	deep	JJ
07-kim-ai	150	19	learning	learn	VBG
07-kim-ai	150	20	techniques.16	techniques.16	NNP
07-kim-ai	150	21	Yewno	Yewno	NNP
07-kim-ai	150	22	Discover	Discover	NNP
07-kim-ai	150	23	displays	display	VBZ
07-kim-ai	150	24	the	the	DT
07-kim-ai	150	25	connections	connection	NNS
07-kim-ai	150	26	between	between	IN
07-kim-ai	150	27	different	different	JJ
07-kim-ai	150	28	con-	con-	NN
07-kim-ai	150	29	cepts	cept	NNS
07-kim-ai	150	30	or	or	CC
07-kim-ai	150	31	works	work	VBZ
07-kim-ai	150	32	in	in	IN
07-kim-ai	150	33	library	library	NN
07-kim-ai	150	34	materials	material	NNS
07-kim-ai	150	35	.	.	.
07-kim-ai	151	1	Quartolio	quartolio	JJ
07-kim-ai	151	2	targets	target	NNS
07-kim-ai	151	3	researchers	researcher	NNS
07-kim-ai	151	4	looking	look	VBG
07-kim-ai	151	5	to	to	TO
07-kim-ai	151	6	discover	discover	VB
07-kim-ai	151	7	untapped	untapped	JJ
07-kim-ai	151	8	research	research	NN
07-kim-ai	151	9	opportunities	opportunity	NNS
07-kim-ai	151	10	based	base	VBN
07-kim-ai	151	11	upon	upon	IN
07-kim-ai	151	12	a	a	DT
07-kim-ai	151	13	large	large	JJ
07-kim-ai	151	14	amount	amount	NN
07-kim-ai	151	15	of	of	IN
07-kim-ai	151	16	data	datum	NNS
07-kim-ai	151	17	that	that	WDT
07-kim-ai	151	18	includes	include	VBZ
07-kim-ai	151	19	articles	article	NNS
07-kim-ai	151	20	,	,	,
07-kim-ai	151	21	clinical	clinical	JJ
07-kim-ai	151	22	trials	trial	NNS
07-kim-ai	151	23	,	,	,
07-kim-ai	151	24	patents	patent	NNS
07-kim-ai	151	25	,	,	,
07-kim-ai	151	26	and	and	CC
07-kim-ai	151	27	notes	note	NNS
07-kim-ai	151	28	.	.	.
07-kim-ai	152	1	Similarly	similarly	RB
07-kim-ai	152	2	,	,	,
07-kim-ai	152	3	Iris.ai	Iris.ai	NNP
07-kim-ai	152	4	helps	help	VBZ
07-kim-ai	152	5	researchers	researcher	NNS
07-kim-ai	152	6	identify	identify	VB
07-kim-ai	152	7	and	and	CC
07-kim-ai	152	8	review	review	VB
07-kim-ai	152	9	a	a	DT
07-kim-ai	152	10	large	large	JJ
07-kim-ai	152	11	amount	amount	NN
07-kim-ai	152	12	of	of	IN
07-kim-ai	152	13	re-	re-	JJ
07-kim-ai	152	14	search	search	NN
07-kim-ai	152	15	papers	paper	NNS
07-kim-ai	152	16	and	and	CC
07-kim-ai	152	17	patents	patent	NNS
07-kim-ai	152	18	and	and	CC
07-kim-ai	152	19	extracts	extract	VBZ
07-kim-ai	152	20	key	key	JJ
07-kim-ai	152	21	information	information	NN
07-kim-ai	152	22	from	from	IN
07-kim-ai	152	23	them	-PRON-	PRP
07-kim-ai	152	24	.	.	.
07-kim-ai	153	1	Kira	Kira	NNP
07-kim-ai	153	2	identifies	identifie	NNS
07-kim-ai	153	3	,	,	,
07-kim-ai	153	4	extracts	extract	VBZ
07-kim-ai	153	5	,	,	,
07-kim-ai	153	6	and	and	CC
07-kim-ai	153	7	analyzes	analyze	VBZ
07-kim-ai	153	8	text	text	NN
07-kim-ai	153	9	in	in	IN
07-kim-ai	153	10	contracts	contract	NNS
07-kim-ai	153	11	and	and	CC
07-kim-ai	153	12	other	other	JJ
07-kim-ai	153	13	legal	legal	JJ
07-kim-ai	153	14	documents.17	documents.17	NNP
07-kim-ai	153	15	None	None	NNP
07-kim-ai	153	16	of	of	IN
07-kim-ai	153	17	these	these	DT
07-kim-ai	153	18	applications	application	NNS
07-kim-ai	153	19	performs	perform	VBZ
07-kim-ai	153	20	fully	fully	RB
07-kim-ai	153	21	automated	automate	VBN
07-kim-ai	153	22	decision	decision	NN
07-kim-ai	153	23	-	-	HYPH
07-kim-ai	153	24	making	making	NN
07-kim-ai	153	25	nor	nor	CC
07-kim-ai	153	26	incorporates	incorporate	VBZ
07-kim-ai	153	27	the	the	DT
07-kim-ai	153	28	digital	digital	JJ
07-kim-ai	153	29	assistant	assistant	NN
07-kim-ai	153	30	feature	feature	NN
07-kim-ai	153	31	.	.	.
07-kim-ai	154	1	But	but	CC
07-kim-ai	154	2	this	this	DT
07-kim-ai	154	3	is	be	VBZ
07-kim-ai	154	4	an	an	DT
07-kim-ai	154	5	area	area	NN
07-kim-ai	154	6	on	on	IN
07-kim-ai	154	7	which	which	WDT
07-kim-ai	154	8	information	information	NN
07-kim-ai	154	9	systems	system	NNS
07-kim-ai	154	10	vendors	vendor	NNS
07-kim-ai	154	11	are	be	VBP
07-kim-ai	154	12	increasingly	increasingly	RB
07-kim-ai	154	13	focusing	focus	VBG
07-kim-ai	154	14	their	-PRON-	PRP$
07-kim-ai	154	15	efforts	effort	NNS
07-kim-ai	154	16	.	.	.
07-kim-ai	155	1	Libraries	library	NNS
07-kim-ai	155	2	themselves	-PRON-	PRP
07-kim-ai	155	3	are	be	VBP
07-kim-ai	155	4	also	also	RB
07-kim-ai	155	5	experimenting	experiment	VBG
07-kim-ai	155	6	with	with	IN
07-kim-ai	155	7	AI	AI	NNP
07-kim-ai	155	8	to	to	TO
07-kim-ai	155	9	test	test	VB
07-kim-ai	155	10	its	-PRON-	PRP$
07-kim-ai	155	11	potential	potential	NN
07-kim-ai	155	12	for	for	IN
07-kim-ai	155	13	library	library	NN
07-kim-ai	155	14	services	service	NNS
07-kim-ai	155	15	and	and	CC
07-kim-ai	155	16	operations	operation	NNS
07-kim-ai	155	17	.	.	.
07-kim-ai	156	1	Some	some	DT
07-kim-ai	156	2	are	be	VBP
07-kim-ai	156	3	focusing	focus	VBG
07-kim-ai	156	4	on	on	IN
07-kim-ai	156	5	using	use	VBG
07-kim-ai	156	6	AI	AI	NNP
07-kim-ai	156	7	,	,	,
07-kim-ai	156	8	particularly	particularly	RB
07-kim-ai	156	9	the	the	DT
07-kim-ai	156	10	voice	voice	NN
07-kim-ai	156	11	user	user	NN
07-kim-ai	156	12	interface	interface	NN
07-kim-ai	156	13	aspect	aspect	NN
07-kim-ai	156	14	of	of	IN
07-kim-ai	156	15	the	the	DT
07-kim-ai	156	16	digital	digital	JJ
07-kim-ai	156	17	assistant	assistant	NN
07-kim-ai	156	18	,	,	,
07-kim-ai	156	19	in	in	IN
07-kim-ai	156	20	order	order	NN
07-kim-ai	156	21	to	to	TO
07-kim-ai	156	22	improve	improve	VB
07-kim-ai	156	23	existing	exist	VBG
07-kim-ai	156	24	services	service	NNS
07-kim-ai	156	25	.	.	.
07-kim-ai	157	1	The	the	DT
07-kim-ai	157	2	University	University	NNP
07-kim-ai	157	3	of	of	IN
07-kim-ai	157	4	Oklahoma	Oklahoma	NNP
07-kim-ai	157	5	Libraries	library	NNS
07-kim-ai	157	6	have	have	VBP
07-kim-ai	157	7	been	be	VBN
07-kim-ai	157	8	building	build	VBG
07-kim-ai	157	9	an	an	DT
07-kim-ai	157	10	Alexa	alexa	JJ
07-kim-ai	157	11	application	application	NN
07-kim-ai	157	12	to	to	TO
07-kim-ai	157	13	provide	provide	VB
07-kim-ai	157	14	basic	basic	JJ
07-kim-ai	157	15	reference	reference	NN
07-kim-ai	157	16	service	service	NN
07-kim-ai	157	17	to	to	IN
07-kim-ai	157	18	their	-PRON-	PRP$
07-kim-ai	157	19	students.18	students.18	NNP
07-kim-ai	157	20	16See	16see	CD
07-kim-ai	157	21	?	?	.
07-kim-ai	157	22	iiTb	iitb	UH
07-kim-ai	157	23	,	,	,
07-kim-ai	157	24	ffrrrXv2rMQX+QKf2	ffrrrxv2rmqx+qkf2	JJ
07-kim-ai	157	25	/	/	NFP
07-kim-ai	157	26	m+	m+	$
07-kim-ai	157	27	�	�	ADD
07-kim-ai	157	28	iBQM	ibqm	NN
07-kim-ai	157	29	,	,	,
07-kim-ai	157	30	?	?	.
07-kim-ai	157	31	iiTb	iitb	UH
07-kim-ai	157	32	,	,	,
07-kim-ai	157	33	ff[m	ff[m	ADD
07-kim-ai	157	34	�	�	NNP
07-kim-ai	157	35	`iQHBQX+QKf	`iqhbqx+qkf	NN
07-kim-ai	157	36	,	,	,
07-kim-ai	157	37	and	and	CC
07-kim-ai	157	38	?	?	.
07-kim-ai	157	39	iiTb	iitb	UH
07-kim-ai	157	40	,	,	,
07-kim-ai	157	41	ffB`BbX	ffB`BbX	NNP
07-kim-ai	157	42	�	�	NNP
07-kim-ai	157	43	Bf	Bf	NNP
07-kim-ai	157	44	.	.	.
07-kim-ai	158	1	17See	17see	LS
07-kim-ai	158	2	?	?	.
07-kim-ai	158	3	iiTb	iitb	UH
07-kim-ai	158	4	,	,	,
07-kim-ai	158	5	ffFB`	ffFB`	NNP
07-kim-ai	158	6	�	�	NNP
07-kim-ai	158	7	bvbi2KbX+QKf	bvbi2KbX+QKf	NNP
07-kim-ai	158	8	.	.	.
07-kim-ai	159	1	Law	law	NN
07-kim-ai	159	2	firms	firm	NNS
07-kim-ai	159	3	are	be	VBP
07-kim-ai	159	4	adopting	adopt	VBG
07-kim-ai	159	5	similar	similar	JJ
07-kim-ai	159	6	products	product	NNS
07-kim-ai	159	7	to	to	TO
07-kim-ai	159	8	automate	automate	VB
07-kim-ai	159	9	and	and	CC
07-kim-ai	159	10	expedite	expedite	VB
07-kim-ai	159	11	their	-PRON-	PRP$
07-kim-ai	159	12	legal	legal	JJ
07-kim-ai	159	13	work	work	NN
07-kim-ai	159	14	,	,	,
07-kim-ai	159	15	and	and	CC
07-kim-ai	159	16	law	law	NN
07-kim-ai	159	17	librarians	librarian	NNS
07-kim-ai	159	18	are	be	VBP
07-kim-ai	159	19	discussing	discuss	VBG
07-kim-ai	159	20	how	how	WRB
07-kim-ai	159	21	the	the	DT
07-kim-ai	159	22	use	use	NN
07-kim-ai	159	23	of	of	IN
07-kim-ai	159	24	AI	AI	NNP
07-kim-ai	159	25	may	may	MD
07-kim-ai	159	26	change	change	VB
07-kim-ai	159	27	their	-PRON-	PRP$
07-kim-ai	159	28	work	work	NN
07-kim-ai	159	29	.	.	.
07-kim-ai	160	1	See	see	VB
07-kim-ai	160	2	Marr	Marr	NNP
07-kim-ai	160	3	2018	2018	CD
07-kim-ai	160	4	and	and	CC
07-kim-ai	160	5	Talley	Talley	NNP
07-kim-ai	160	6	2016	2016	CD
07-kim-ai	160	7	.	.	.
07-kim-ai	161	1	18University	18university	CD
07-kim-ai	161	2	of	of	IN
07-kim-ai	161	3	Oklahoma	Oklahoma	NNP
07-kim-ai	161	4	Libraries	Libraries	NNPS
07-kim-ai	161	5	are	be	VBP
07-kim-ai	161	6	building	build	VBG
07-kim-ai	161	7	an	an	DT
07-kim-ai	161	8	Alexa	alexa	JJ
07-kim-ai	161	9	application	application	NN
07-kim-ai	161	10	that	that	WDT
07-kim-ai	161	11	will	will	MD
07-kim-ai	161	12	provide	provide	VB
07-kim-ai	161	13	some	some	DT
07-kim-ai	161	14	basic	basic	JJ
07-kim-ai	161	15	reference	reference	NN
07-kim-ai	161	16	service	service	NN
07-kim-ai	161	17	to	to	IN
07-kim-ai	161	18	their	-PRON-	PRP$
07-kim-ai	161	19	students	student	NNS
07-kim-ai	161	20	.	.	.
07-kim-ai	162	1	Also	also	RB
07-kim-ai	162	2	,	,	,
07-kim-ai	162	3	their	-PRON-	PRP$
07-kim-ai	162	4	PAIR	PAIR	NNP
07-kim-ai	162	5	registry	registry	NN
07-kim-ai	162	6	attempts	attempt	VBZ
07-kim-ai	162	7	to	to	TO
07-kim-ai	162	8	compile	compile	VB
07-kim-ai	162	9	all	all	DT
07-kim-ai	162	10	AI	ai	NN
07-kim-ai	162	11	-	-	HYPH
07-kim-ai	162	12	related	relate	VBN
07-kim-ai	162	13	projects	project	NNS
07-kim-ai	162	14	at	at	IN
07-kim-ai	162	15	libraries	library	NNS
07-kim-ai	162	16	.	.	.
07-kim-ai	163	1	See	see	VB
07-kim-ai	163	2	?	?	.
07-kim-ai	163	3	iiTb	iitb	UH
07-kim-ai	163	4	,	,	,
07-kim-ai	163	5	ffT	ffT	NNP
07-kim-ai	163	6	�	�	NNP
07-kim-ai	163	7	B`XH	B`XH	NNP
07-kim-ai	163	8	B#`	B#`	NNP
07-kim-ai	163	9	�	�	NNP
07-kim-ai	163	10	`B2bXQmX2	`B2bXQmX2	.
07-kim-ai	163	11	/	/	SYM
07-kim-ai	163	12	m	m	NN
07-kim-ai	163	13	.	.	.
07-kim-ai	164	1	https://www.yewno.com/education	https://www.yewno.com/education	ADD
07-kim-ai	164	2	https://quartolio.com/	https://quartolio.com/	CD
07-kim-ai	164	3	https://iris.ai/	https://iris.ai/	NNP
07-kim-ai	164	4	https://kirasystems.com/	https://kirasystems.com/	CD
07-kim-ai	164	5	https://pair.libraries.ou.edu	https://pair.libraries.ou.edu	NN
07-kim-ai	164	6	https://pair.libraries.ou.edu	https://pair.libraries.ou.edu	NFP
07-kim-ai	164	7	Kim	Kim	NNP
07-kim-ai	164	8	79	79	CD
07-kim-ai	164	9	At	at	IN
07-kim-ai	164	10	the	the	DT
07-kim-ai	164	11	University	University	NNP
07-kim-ai	164	12	of	of	IN
07-kim-ai	164	13	Pretoria	Pretoria	NNP
07-kim-ai	164	14	Library	Library	NNP
07-kim-ai	164	15	in	in	IN
07-kim-ai	164	16	South	South	NNP
07-kim-ai	164	17	Africa	Africa	NNP
07-kim-ai	164	18	,	,	,
07-kim-ai	164	19	a	a	DT
07-kim-ai	164	20	robot	robot	NN
07-kim-ai	164	21	named	name	VBN
07-kim-ai	164	22	‘	'	``
07-kim-ai	164	23	Libby	Libby	NNP
07-kim-ai	164	24	’	'	''
07-kim-ai	164	25	already	already	RB
07-kim-ai	164	26	interacts	interact	VBZ
07-kim-ai	164	27	with	with	IN
07-kim-ai	164	28	patrons	patron	NNS
07-kim-ai	164	29	by	by	IN
07-kim-ai	164	30	providing	provide	VBG
07-kim-ai	164	31	guidance	guidance	NN
07-kim-ai	164	32	,	,	,
07-kim-ai	164	33	answering	answer	VBG
07-kim-ai	164	34	questions	question	NNS
07-kim-ai	164	35	,	,	,
07-kim-ai	164	36	conducting	conduct	VBG
07-kim-ai	164	37	surveys	survey	NNS
07-kim-ai	164	38	,	,	,
07-kim-ai	164	39	and	and	CC
07-kim-ai	164	40	displaying	display	VBG
07-kim-ai	164	41	marketing	marketing	NN
07-kim-ai	164	42	videos	video	NNS
07-kim-ai	164	43	(	(	-LRB-
07-kim-ai	164	44	Mahlangu	Mahlangu	NNP
07-kim-ai	164	45	2019	2019	CD
07-kim-ai	164	46	)	)	-RRB-
07-kim-ai	164	47	.	.	.
07-kim-ai	165	1	Other	other	JJ
07-kim-ai	165	2	libraries	library	NNS
07-kim-ai	165	3	are	be	VBP
07-kim-ai	165	4	applying	apply	VBG
07-kim-ai	165	5	AI	AI	NNP
07-kim-ai	165	6	to	to	TO
07-kim-ai	165	7	extract	extract	VB
07-kim-ai	165	8	information	information	NN
07-kim-ai	165	9	from	from	IN
07-kim-ai	165	10	digital	digital	JJ
07-kim-ai	165	11	materials	material	NNS
07-kim-ai	165	12	and	and	CC
07-kim-ai	165	13	automate	automate	VB
07-kim-ai	165	14	metadata	metadata	NN
07-kim-ai	165	15	generation	generation	NN
07-kim-ai	165	16	to	to	TO
07-kim-ai	165	17	enhance	enhance	VB
07-kim-ai	165	18	their	-PRON-	PRP$
07-kim-ai	165	19	discovery	discovery	NN
07-kim-ai	165	20	and	and	CC
07-kim-ai	165	21	use	use	NN
07-kim-ai	165	22	.	.	.
07-kim-ai	166	1	The	the	DT
07-kim-ai	166	2	Library	Library	NNP
07-kim-ai	166	3	of	of	IN
07-kim-ai	166	4	Congress	Congress	NNP
07-kim-ai	166	5	has	have	VBZ
07-kim-ai	166	6	worked	work	VBN
07-kim-ai	166	7	on	on	IN
07-kim-ai	166	8	detecting	detect	VBG
07-kim-ai	166	9	features	feature	NNS
07-kim-ai	166	10	,	,	,
07-kim-ai	166	11	such	such	JJ
07-kim-ai	166	12	as	as	IN
07-kim-ai	166	13	railroads	railroad	NNS
07-kim-ai	166	14	in	in	IN
07-kim-ai	166	15	maps	map	NNS
07-kim-ai	166	16	,	,	,
07-kim-ai	166	17	using	use	VBG
07-kim-ai	166	18	the	the	DT
07-kim-ai	166	19	convolutional	convolutional	JJ
07-kim-ai	166	20	neural	neural	JJ
07-kim-ai	166	21	network	network	NN
07-kim-ai	166	22	model	model	NN
07-kim-ai	166	23	,	,	,
07-kim-ai	166	24	and	and	CC
07-kim-ai	166	25	issued	issue	VBD
07-kim-ai	166	26	a	a	DT
07-kim-ai	166	27	solicitation	solicitation	NN
07-kim-ai	166	28	for	for	IN
07-kim-ai	166	29	a	a	DT
07-kim-ai	166	30	machine	machine	NN
07-kim-ai	166	31	learning	learning	NN
07-kim-ai	166	32	and	and	CC
07-kim-ai	166	33	deep	deep	JJ
07-kim-ai	166	34	learning	learning	NN
07-kim-ai	166	35	pilot	pilot	NN
07-kim-ai	166	36	program	program	NN
07-kim-ai	166	37	that	that	WDT
07-kim-ai	166	38	will	will	MD
07-kim-ai	166	39	max-	max-	RB
07-kim-ai	166	40	imize	imize	VB
07-kim-ai	166	41	the	the	DT
07-kim-ai	166	42	use	use	NN
07-kim-ai	166	43	of	of	IN
07-kim-ai	166	44	its	-PRON-	PRP$
07-kim-ai	166	45	digital	digital	JJ
07-kim-ai	166	46	collections	collection	NNS
07-kim-ai	166	47	in	in	IN
07-kim-ai	166	48	2019.19	2019.19	CD
07-kim-ai	166	49	Indiana	Indiana	NNP
07-kim-ai	166	50	University	University	NNP
07-kim-ai	166	51	Libraries	Libraries	NNPS
07-kim-ai	166	52	,	,	,
07-kim-ai	166	53	AVP	AVP	NNP
07-kim-ai	166	54	,	,	,
07-kim-ai	166	55	University	University	NNP
07-kim-ai	166	56	of	of	IN
07-kim-ai	166	57	Texas	Texas	NNP
07-kim-ai	166	58	Austin	Austin	NNP
07-kim-ai	166	59	School	School	NNP
07-kim-ai	166	60	of	of	IN
07-kim-ai	166	61	Information	Information	NNP
07-kim-ai	166	62	,	,	,
07-kim-ai	166	63	and	and	CC
07-kim-ai	166	64	the	the	DT
07-kim-ai	166	65	New	New	NNP
07-kim-ai	166	66	York	York	NNP
07-kim-ai	166	67	Public	Public	NNP
07-kim-ai	166	68	Library	Library	NNP
07-kim-ai	166	69	are	be	VBP
07-kim-ai	166	70	jointly	jointly	RB
07-kim-ai	166	71	developing	develop	VBG
07-kim-ai	166	72	the	the	DT
07-kim-ai	166	73	Audiovisual	Audiovisual	NNP
07-kim-ai	166	74	Metadata	Metadata	NNP
07-kim-ai	166	75	Platform	Platform	NNP
07-kim-ai	166	76	(	(	-LRB-
07-kim-ai	166	77	AMP	AMP	NNP
07-kim-ai	166	78	)	)	-RRB-
07-kim-ai	166	79	,	,	,
07-kim-ai	166	80	using	use	VBG
07-kim-ai	166	81	many	many	JJ
07-kim-ai	166	82	AI	ai	NN
07-kim-ai	166	83	tools	tool	NNS
07-kim-ai	166	84	in	in	IN
07-kim-ai	166	85	order	order	NN
07-kim-ai	166	86	to	to	TO
07-kim-ai	166	87	automatically	automatically	RB
07-kim-ai	166	88	gen-	gen-	VB
07-kim-ai	166	89	erate	erate	VB
07-kim-ai	166	90	metadata	metadata	NN
07-kim-ai	166	91	for	for	IN
07-kim-ai	166	92	audiovisual	audiovisual	JJ
07-kim-ai	166	93	materials	material	NNS
07-kim-ai	166	94	,	,	,
07-kim-ai	166	95	which	which	WDT
07-kim-ai	166	96	collection	collection	NN
07-kim-ai	166	97	managers	manager	NNS
07-kim-ai	166	98	can	can	MD
07-kim-ai	166	99	use	use	VB
07-kim-ai	166	100	to	to	TO
07-kim-ai	166	101	supplement	supplement	VB
07-kim-ai	166	102	their	-PRON-	PRP$
07-kim-ai	166	103	archival	archival	NN
07-kim-ai	166	104	description	description	NN
07-kim-ai	166	105	and	and	CC
07-kim-ai	166	106	processing	processing	NN
07-kim-ai	166	107	workflows.20	workflows.20	NNP
07-kim-ai	166	108	Some	some	DT
07-kim-ai	166	109	libraries	library	NNS
07-kim-ai	166	110	are	be	VBP
07-kim-ai	166	111	also	also	RB
07-kim-ai	166	112	testing	test	VBG
07-kim-ai	166	113	out	out	RP
07-kim-ai	166	114	AI	AI	NNP
07-kim-ai	166	115	as	as	IN
07-kim-ai	166	116	a	a	DT
07-kim-ai	166	117	tool	tool	NN
07-kim-ai	166	118	for	for	IN
07-kim-ai	166	119	evaluating	evaluate	VBG
07-kim-ai	166	120	services	service	NNS
07-kim-ai	166	121	and	and	CC
07-kim-ai	166	122	operations	operation	NNS
07-kim-ai	166	123	.	.	.
07-kim-ai	167	1	The	the	DT
07-kim-ai	167	2	Uni-	Uni-	NNP
07-kim-ai	167	3	versity	versity	NN
07-kim-ai	167	4	of	of	IN
07-kim-ai	167	5	Rochester	Rochester	NNP
07-kim-ai	167	6	Libraries	Libraries	NNP
07-kim-ai	167	7	applied	apply	VBD
07-kim-ai	167	8	deep	deep	JJ
07-kim-ai	167	9	learning	learning	NN
07-kim-ai	167	10	to	to	IN
07-kim-ai	167	11	the	the	DT
07-kim-ai	167	12	library	library	NN
07-kim-ai	167	13	’s	’s	POS
07-kim-ai	167	14	space	space	NN
07-kim-ai	167	15	assessment	assessment	NN
07-kim-ai	167	16	to	to	TO
07-kim-ai	167	17	determine	determine	VB
07-kim-ai	167	18	the	the	DT
07-kim-ai	167	19	optimal	optimal	JJ
07-kim-ai	167	20	staffing	staffing	NN
07-kim-ai	167	21	level	level	NN
07-kim-ai	167	22	and	and	CC
07-kim-ai	167	23	building	building	NN
07-kim-ai	167	24	hours	hour	NNS
07-kim-ai	167	25	.	.	.
07-kim-ai	168	1	The	the	DT
07-kim-ai	168	2	University	University	NNP
07-kim-ai	168	3	of	of	IN
07-kim-ai	168	4	Illinois	Illinois	NNP
07-kim-ai	168	5	Urbana	Urbana	NNP
07-kim-ai	168	6	-	-	HYPH
07-kim-ai	168	7	Champaign	Champaign	NNP
07-kim-ai	168	8	Li-	Li-	NNP
07-kim-ai	168	9	braries	brarie	NNS
07-kim-ai	168	10	used	use	VBD
07-kim-ai	168	11	machine	machine	NN
07-kim-ai	168	12	learning	learn	VBG
07-kim-ai	168	13	to	to	TO
07-kim-ai	168	14	conduct	conduct	VB
07-kim-ai	168	15	sentiment	sentiment	NN
07-kim-ai	168	16	analysis	analysis	NN
07-kim-ai	168	17	on	on	IN
07-kim-ai	168	18	their	-PRON-	PRP$
07-kim-ai	168	19	reference	reference	NN
07-kim-ai	168	20	chat	chat	NN
07-kim-ai	168	21	log	log	NN
07-kim-ai	168	22	(	(	-LRB-
07-kim-ai	168	23	Blewer	Blewer	NNP
07-kim-ai	168	24	,	,	,
07-kim-ai	168	25	Kim	Kim	NNP
07-kim-ai	168	26	,	,	,
07-kim-ai	168	27	and	and	CC
07-kim-ai	168	28	Phetteplace	Phetteplace	NNP
07-kim-ai	168	29	2018	2018	CD
07-kim-ai	168	30	)	)	-RRB-
07-kim-ai	168	31	.	.	.
07-kim-ai	169	1	Ethical	ethical	JJ
07-kim-ai	169	2	Challenges	Challenges	NNPS
07-kim-ai	169	3	from	from	IN
07-kim-ai	169	4	the	the	DT
07-kim-ai	169	5	Personalized	Personalized	NNP
07-kim-ai	169	6	and	and	CC
07-kim-ai	169	7	Automated	Automated	NNP
07-kim-ai	169	8	Information	Information	NNP
07-kim-ai	169	9	Environment	Environment	NNP
07-kim-ai	169	10	Do	do	VBP
07-kim-ai	169	11	these	these	DT
07-kim-ai	169	12	current	current	JJ
07-kim-ai	169	13	and	and	CC
07-kim-ai	169	14	future	future	JJ
07-kim-ai	169	15	AI	AI	NNP
07-kim-ai	169	16	applications	application	NNS
07-kim-ai	169	17	for	for	IN
07-kim-ai	169	18	libraries	library	NNS
07-kim-ai	169	19	pose	pose	VBP
07-kim-ai	169	20	ethical	ethical	JJ
07-kim-ai	169	21	challenges	challenge	NNS
07-kim-ai	169	22	similar	similar	JJ
07-kim-ai	169	23	to	to	IN
07-kim-ai	169	24	those	those	DT
07-kim-ai	169	25	that	that	WDT
07-kim-ai	169	26	we	-PRON-	PRP
07-kim-ai	169	27	discussed	discuss	VBD
07-kim-ai	169	28	earlier	early	RBR
07-kim-ai	169	29	?	?	.
07-kim-ai	170	1	Since	since	IN
07-kim-ai	170	2	information	information	NN
07-kim-ai	170	3	query	query	NN
07-kim-ai	170	4	,	,	,
07-kim-ai	170	5	discovery	discovery	NN
07-kim-ai	170	6	,	,	,
07-kim-ai	170	7	and	and	CC
07-kim-ai	170	8	retrieval	retrieval	NN
07-kim-ai	170	9	rarely	rarely	RB
07-kim-ai	170	10	involve	involve	VBP
07-kim-ai	170	11	life-	life-	JJ
07-kim-ai	170	12	or	or	CC
07-kim-ai	170	13	-	-	HYPH
07-kim-ai	170	14	death	death	NN
07-kim-ai	170	15	situations	situation	NNS
07-kim-ai	170	16	,	,	,
07-kim-ai	170	17	stakes	stake	NNS
07-kim-ai	170	18	seem	seem	VBP
07-kim-ai	170	19	to	to	TO
07-kim-ai	170	20	be	be	VB
07-kim-ai	170	21	certainly	certainly	RB
07-kim-ai	170	22	lower	low	JJR
07-kim-ai	170	23	.	.	.
07-kim-ai	171	1	But	but	CC
07-kim-ai	171	2	an	an	DT
07-kim-ai	171	3	AI	ai	NN
07-kim-ai	171	4	-	-	HYPH
07-kim-ai	171	5	driven	drive	VBN
07-kim-ai	171	6	automated	automated	JJ
07-kim-ai	171	7	information	information	NN
07-kim-ai	171	8	environment	environment	NN
07-kim-ai	171	9	does	do	VBZ
07-kim-ai	171	10	raise	raise	VB
07-kim-ai	171	11	its	-PRON-	PRP$
07-kim-ai	171	12	own	own	JJ
07-kim-ai	171	13	distinct	distinct	JJ
07-kim-ai	171	14	ethical	ethical	JJ
07-kim-ai	171	15	challenges	challenge	NNS
07-kim-ai	171	16	.	.	.
07-kim-ai	172	1	(	(	-LRB-
07-kim-ai	172	2	i	i	NN
07-kim-ai	172	3	)	)	-RRB-
07-kim-ai	172	4	Intellectual	intellectual	JJ
07-kim-ai	172	5	isolation	isolation	NN
07-kim-ai	172	6	and	and	CC
07-kim-ai	172	7	bigotry	bigotry	NN
07-kim-ai	172	8	hampering	hamper	VBG
07-kim-ai	172	9	civic	civic	JJ
07-kim-ai	172	10	discourse	discourse	NN
07-kim-ai	172	11	Many	many	JJ
07-kim-ai	172	12	AI	AI	NNP
07-kim-ai	172	13	applications	application	NNS
07-kim-ai	172	14	that	that	WDT
07-kim-ai	172	15	assist	assist	VBP
07-kim-ai	172	16	with	with	IN
07-kim-ai	172	17	information	information	NN
07-kim-ai	172	18	seeking	seek	VBG
07-kim-ai	172	19	activities	activity	NNS
07-kim-ai	172	20	promise	promise	VBP
07-kim-ai	172	21	a	a	DT
07-kim-ai	172	22	higher	high	JJR
07-kim-ai	172	23	level	level	NN
07-kim-ai	172	24	of	of	IN
07-kim-ai	172	25	per-	per-	DT
07-kim-ai	172	26	sonalization	sonalization	NN
07-kim-ai	172	27	.	.	.
07-kim-ai	173	1	But	but	CC
07-kim-ai	173	2	a	a	DT
07-kim-ai	173	3	highly	highly	RB
07-kim-ai	173	4	personalized	personalized	JJ
07-kim-ai	173	5	information	information	NN
07-kim-ai	173	6	environment	environment	NN
07-kim-ai	173	7	often	often	RB
07-kim-ai	173	8	traps	trap	VBZ
07-kim-ai	173	9	people	people	NNS
07-kim-ai	173	10	in	in	IN
07-kim-ai	173	11	their	-PRON-	PRP$
07-kim-ai	173	12	own	own	JJ
07-kim-ai	173	13	so	so	RB
07-kim-ai	173	14	-	-	HYPH
07-kim-ai	173	15	called	call	VBN
07-kim-ai	173	16	‘	'	``
07-kim-ai	173	17	filter	filter	NN
07-kim-ai	173	18	bubble	bubble	NN
07-kim-ai	173	19	,	,	,
07-kim-ai	173	20	’	'	''
07-kim-ai	173	21	as	as	IN
07-kim-ai	173	22	we	-PRON-	PRP
07-kim-ai	173	23	have	have	VBP
07-kim-ai	173	24	been	be	VBN
07-kim-ai	173	25	increasingly	increasingly	RB
07-kim-ai	173	26	seeing	see	VBG
07-kim-ai	173	27	in	in	IN
07-kim-ai	173	28	today	today	NN
07-kim-ai	173	29	’s	’s	POS
07-kim-ai	173	30	social	social	JJ
07-kim-ai	173	31	media	medium	NNS
07-kim-ai	173	32	channels	channel	NNS
07-kim-ai	173	33	,	,	,
07-kim-ai	173	34	news	news	NN
07-kim-ai	173	35	websites	website	NNS
07-kim-ai	173	36	,	,	,
07-kim-ai	173	37	and	and	CC
07-kim-ai	173	38	commercial	commercial	JJ
07-kim-ai	173	39	search	search	NN
07-kim-ai	173	40	engines	engine	NNS
07-kim-ai	173	41	,	,	,
07-kim-ai	173	42	where	where	WRB
07-kim-ai	173	43	such	such	JJ
07-kim-ai	173	44	personalization	personalization	NN
07-kim-ai	173	45	is	be	VBZ
07-kim-ai	173	46	provided	provide	VBN
07-kim-ai	173	47	by	by	IN
07-kim-ai	173	48	machine	machine	NN
07-kim-ai	173	49	learning	learning	NN
07-kim-ai	173	50	and	and	CC
07-kim-ai	173	51	deep	deep	JJ
07-kim-ai	173	52	learning.21	learning.21	NNP
07-kim-ai	173	53	Sophisticated	Sophisticated	NNP
07-kim-ai	173	54	AI	AI	NNP
07-kim-ai	173	55	algorithms	algorithm	NNS
07-kim-ai	173	56	are	be	VBP
07-kim-ai	173	57	already	already	RB
07-kim-ai	173	58	curating	curate	VBG
07-kim-ai	173	59	and	and	CC
07-kim-ai	173	60	pushing	push	VBG
07-kim-ai	173	61	in-	in-	JJ
07-kim-ai	173	62	formation	formation	NN
07-kim-ai	173	63	feeds	feed	VBZ
07-kim-ai	173	64	based	base	VBN
07-kim-ai	173	65	upon	upon	IN
07-kim-ai	173	66	the	the	DT
07-kim-ai	173	67	person	person	NN
07-kim-ai	173	68	’s	’s	POS
07-kim-ai	173	69	past	past	JJ
07-kim-ai	173	70	search	search	NN
07-kim-ai	173	71	and	and	CC
07-kim-ai	173	72	click	click	VB
07-kim-ai	173	73	behavior	behavior	NN
07-kim-ai	173	74	.	.	.
07-kim-ai	174	1	The	the	DT
07-kim-ai	174	2	result	result	NN
07-kim-ai	174	3	is	be	VBZ
07-kim-ai	174	4	that	that	IN
07-kim-ai	174	5	infor-	infor-	JJ
07-kim-ai	174	6	mation	mation	NN
07-kim-ai	174	7	seekers	seeker	NNS
07-kim-ai	174	8	are	be	VBP
07-kim-ai	174	9	provided	provide	VBN
07-kim-ai	174	10	with	with	IN
07-kim-ai	174	11	information	information	NN
07-kim-ai	174	12	that	that	WDT
07-kim-ai	174	13	conforms	conform	VBZ
07-kim-ai	174	14	and	and	CC
07-kim-ai	174	15	reinforces	reinforce	VBZ
07-kim-ai	174	16	their	-PRON-	PRP$
07-kim-ai	174	17	existing	exist	VBG
07-kim-ai	174	18	beliefs	belief	NNS
07-kim-ai	174	19	and	and	CC
07-kim-ai	174	20	interests	interest	NNS
07-kim-ai	174	21	.	.	.
07-kim-ai	175	1	Views	view	NNS
07-kim-ai	175	2	that	that	WDT
07-kim-ai	175	3	are	be	VBP
07-kim-ai	175	4	novel	novel	JJ
07-kim-ai	175	5	or	or	CC
07-kim-ai	175	6	contrast	contrast	NN
07-kim-ai	175	7	with	with	IN
07-kim-ai	175	8	their	-PRON-	PRP$
07-kim-ai	175	9	existing	exist	VBG
07-kim-ai	175	10	beliefs	belief	NNS
07-kim-ai	175	11	are	be	VBP
07-kim-ai	175	12	suppressed	suppress	VBN
07-kim-ai	175	13	and	and	CC
07-kim-ai	175	14	be-	be-	XX
07-kim-ai	175	15	come	come	VB
07-kim-ai	175	16	invisible	invisible	JJ
07-kim-ai	175	17	without	without	IN
07-kim-ai	175	18	them	-PRON-	PRP
07-kim-ai	175	19	even	even	RB
07-kim-ai	175	20	realizing	realize	VBG
07-kim-ai	175	21	.	.	.
07-kim-ai	176	1	Such	such	JJ
07-kim-ai	176	2	lack	lack	NN
07-kim-ai	176	3	of	of	IN
07-kim-ai	176	4	exposure	exposure	NN
07-kim-ai	176	5	to	to	IN
07-kim-ai	176	6	opposing	oppose	VBG
07-kim-ai	176	7	views	view	NNS
07-kim-ai	176	8	leads	lead	VBZ
07-kim-ai	176	9	information	information	NN
07-kim-ai	176	10	users	user	NNS
07-kim-ai	176	11	to	to	IN
07-kim-ai	176	12	intellectual	intellectual	JJ
07-kim-ai	176	13	isolation	isolation	NN
07-kim-ai	176	14	and	and	CC
07-kim-ai	176	15	even	even	RB
07-kim-ai	176	16	bigotry	bigotry	NN
07-kim-ai	176	17	.	.	.
07-kim-ai	177	1	Highly	highly	RB
07-kim-ai	177	2	personalized	personalized	JJ
07-kim-ai	177	3	information	information	NN
07-kim-ai	177	4	environments	environment	NNS
07-kim-ai	177	5	powered	power	VBN
07-kim-ai	177	6	by	by	IN
07-kim-ai	177	7	AI	AI	NNP
07-kim-ai	177	8	can	can	MD
07-kim-ai	177	9	actively	actively	RB
07-kim-ai	177	10	restrict	restrict	VB
07-kim-ai	177	11	ways	way	NNS
07-kim-ai	177	12	in	in	IN
07-kim-ai	177	13	which	which	WDT
07-kim-ai	177	14	people	people	NNS
07-kim-ai	177	15	develop	develop	VBP
07-kim-ai	177	16	balanced	balanced	JJ
07-kim-ai	177	17	and	and	CC
07-kim-ai	177	18	informed	informed	JJ
07-kim-ai	177	19	opinions	opinion	NNS
07-kim-ai	177	20	,	,	,
07-kim-ai	177	21	thereby	thereby	RB
07-kim-ai	177	22	intensifying	intensify	VBG
07-kim-ai	177	23	and	and	CC
07-kim-ai	177	24	perpet-	perpet-	XX
07-kim-ai	177	25	uating	uate	VBG
07-kim-ai	177	26	social	social	JJ
07-kim-ai	177	27	discord	discord	NN
07-kim-ai	177	28	and	and	CC
07-kim-ai	177	29	disrupting	disrupt	VBG
07-kim-ai	177	30	civic	civic	JJ
07-kim-ai	177	31	discourse	discourse	NN
07-kim-ai	177	32	.	.	.
07-kim-ai	178	1	Under	under	IN
07-kim-ai	178	2	such	such	JJ
07-kim-ai	178	3	conditions	condition	NNS
07-kim-ai	178	4	,	,	,
07-kim-ai	178	5	prejudices	prejudice	NNS
07-kim-ai	178	6	,	,	,
07-kim-ai	178	7	discrim-	discrim-	VBZ
07-kim-ai	178	8	19See	19see	CD
07-kim-ai	178	9	Blewer	Blewer	NNP
07-kim-ai	178	10	,	,	,
07-kim-ai	178	11	Kim	Kim	NNP
07-kim-ai	178	12	,	,	,
07-kim-ai	178	13	and	and	CC
07-kim-ai	178	14	Phetteplace	Phetteplace	NNP
07-kim-ai	178	15	2018	2018	CD
07-kim-ai	178	16	and	and	CC
07-kim-ai	178	17	Price	Price	NNP
07-kim-ai	178	18	2019	2019	CD
07-kim-ai	178	19	.	.	.
07-kim-ai	179	1	20The	20the	CD
07-kim-ai	179	2	AMP	AMP	NNP
07-kim-ai	179	3	wiki	wiki	NN
07-kim-ai	179	4	is	be	VBZ
07-kim-ai	179	5	?	?	.
07-kim-ai	179	6	iiTb	iitb	UH
07-kim-ai	179	7	,	,	,
07-kim-ai	179	8	ffrBFBX	ffrBFBX	NNP
07-kim-ai	179	9	/	/	SYM
07-kim-ai	179	10	HB#XBM	HB#XBM	NNS
07-kim-ai	179	11	/	/	SYM
07-kim-ai	179	12	B	B	NNP
07-kim-ai	179	13	�	�	NNP
07-kim-ai	179	14	M	M	NNP
07-kim-ai	179	15	�	�	NNP
07-kim-ai	179	16	X2	x2	NN
07-kim-ai	179	17	/	/	SYM
07-kim-ai	179	18	mfT	mfT	NNP
07-kim-ai	179	19	�	�	NNS
07-kim-ai	179	20	;2bfpB2rT	;2bfpB2rT	:
07-kim-ai	179	21	�	�	NNP
07-kim-ai	179	22	;2X	;2X	NNP
07-kim-ai	179	23	�	�	NNP
07-kim-ai	179	24	+iBQM?T	+ibqm?t	ADD
07-kim-ai	179	25	�	�	NNP
07-kim-ai	179	26	;2A/48jReNNN9R.	;2A/48jReNNN9R.	.
07-kim-ai	180	1	The	the	DT
07-kim-ai	180	2	Audiovisual	audiovisual	JJ
07-kim-ai	180	3	Metadata	metadata	NN
07-kim-ai	180	4	Platform	platform	NN
07-kim-ai	180	5	Pilot	pilot	NN
07-kim-ai	180	6	Development	Development	NNP
07-kim-ai	180	7	(	(	-LRB-
07-kim-ai	180	8	AMPPD	AMPPD	NNP
07-kim-ai	180	9	)	)	-RRB-
07-kim-ai	180	10	project	project	NN
07-kim-ai	180	11	was	be	VBD
07-kim-ai	180	12	presented	present	VBN
07-kim-ai	180	13	at	at	IN
07-kim-ai	180	14	Code4Lib	Code4Lib	NNP
07-kim-ai	180	15	2020	2020	CD
07-kim-ai	180	16	(	(	-LRB-
07-kim-ai	180	17	Averkamp	Averkamp	NNP
07-kim-ai	180	18	and	and	CC
07-kim-ai	180	19	Hardesty	Hardesty	NNP
07-kim-ai	180	20	2020	2020	CD
07-kim-ai	180	21	)	)	-RRB-
07-kim-ai	180	22	.	.	.
07-kim-ai	181	1	21See	21see	CD
07-kim-ai	181	2	Pariser	Pariser	NNP
07-kim-ai	181	3	2012	2012	CD
07-kim-ai	181	4	.	.	.
07-kim-ai	181	5	https://wiki.dlib.indiana.edu/pages/viewpage.action?pageId=531699941	https://wiki.dlib.indiana.edu/pages/viewpage.action?pageid=531699941	CD
07-kim-ai	181	6	80	80	CD
07-kim-ai	181	7	Machine	Machine	NNP
07-kim-ai	181	8	Learning	Learning	NNP
07-kim-ai	181	9	,	,	,
07-kim-ai	181	10	Libraries	Libraries	NNPS
07-kim-ai	181	11	,	,	,
07-kim-ai	181	12	and	and	CC
07-kim-ai	181	13	Cross	Cross	NNP
07-kim-ai	181	14	-	-	NNP
07-kim-ai	181	15	Disciplinary	Disciplinary	NNP
07-kim-ai	181	16	ResearchǔChapter	ResearchǔChapter	NNP
07-kim-ai	181	17	7	7	CD
07-kim-ai	181	18	ination	ination	NN
07-kim-ai	181	19	,	,	,
07-kim-ai	181	20	and	and	CC
07-kim-ai	181	21	other	other	JJ
07-kim-ai	181	22	unjust	unjust	JJ
07-kim-ai	181	23	social	social	JJ
07-kim-ai	181	24	practices	practice	NNS
07-kim-ai	181	25	are	be	VBP
07-kim-ai	181	26	likely	likely	JJ
07-kim-ai	181	27	to	to	TO
07-kim-ai	181	28	increase	increase	VB
07-kim-ai	181	29	,	,	,
07-kim-ai	181	30	and	and	CC
07-kim-ai	181	31	this	this	DT
07-kim-ai	181	32	in	in	IN
07-kim-ai	181	33	turn	turn	NN
07-kim-ai	181	34	will	will	MD
07-kim-ai	181	35	have	have	VB
07-kim-ai	181	36	more	more	JJR
07-kim-ai	181	37	negative	negative	JJ
07-kim-ai	181	38	impact	impact	NN
07-kim-ai	181	39	on	on	IN
07-kim-ai	181	40	those	those	DT
07-kim-ai	181	41	with	with	IN
07-kim-ai	181	42	fewer	few	JJR
07-kim-ai	181	43	privileges	privilege	NNS
07-kim-ai	181	44	.	.	.
07-kim-ai	182	1	Intellectual	intellectual	JJ
07-kim-ai	182	2	isolation	isolation	NN
07-kim-ai	182	3	and	and	CC
07-kim-ai	182	4	bigotry	bigotry	NN
07-kim-ai	182	5	has	have	VBZ
07-kim-ai	182	6	a	a	DT
07-kim-ai	182	7	distinctly	distinctly	RB
07-kim-ai	182	8	moral	moral	JJ
07-kim-ai	182	9	impact	impact	NN
07-kim-ai	182	10	on	on	IN
07-kim-ai	182	11	society	society	NN
07-kim-ai	182	12	.	.	.
07-kim-ai	183	1	(	(	-LRB-
07-kim-ai	183	2	ii	ii	NNP
07-kim-ai	183	3	)	)	-RRB-
07-kim-ai	183	4	Weakening	Weakening	NNP
07-kim-ai	183	5	of	of	IN
07-kim-ai	183	6	cognitive	cognitive	JJ
07-kim-ai	183	7	agency	agency	NN
07-kim-ai	183	8	and	and	CC
07-kim-ai	183	9	autonomy	autonomy	NN
07-kim-ai	183	10	We	-PRON-	PRP
07-kim-ai	183	11	have	have	VBP
07-kim-ai	183	12	seen	see	VBN
07-kim-ai	183	13	earlier	early	RBR
07-kim-ai	183	14	that	that	IN
07-kim-ai	183	15	AI	AI	NNP
07-kim-ai	183	16	-	-	HYPH
07-kim-ai	183	17	powered	power	VBN
07-kim-ai	183	18	digital	digital	JJ
07-kim-ai	183	19	assistants	assistant	NNS
07-kim-ai	183	20	are	be	VBP
07-kim-ai	183	21	likely	likely	JJ
07-kim-ai	183	22	to	to	TO
07-kim-ai	183	23	mediate	mediate	VB
07-kim-ai	183	24	people	people	NNS
07-kim-ai	183	25	’s	’s	POS
07-kim-ai	183	26	information	information	NN
07-kim-ai	183	27	search	search	NN
07-kim-ai	183	28	,	,	,
07-kim-ai	183	29	discovery	discovery	NN
07-kim-ai	183	30	,	,	,
07-kim-ai	183	31	and	and	CC
07-kim-ai	183	32	retrieval	retrieval	NN
07-kim-ai	183	33	activities	activity	NNS
07-kim-ai	183	34	in	in	IN
07-kim-ai	183	35	the	the	DT
07-kim-ai	183	36	near	near	JJ
07-kim-ai	183	37	future	future	NN
07-kim-ai	183	38	.	.	.
07-kim-ai	184	1	As	as	IN
07-kim-ai	184	2	those	those	DT
07-kim-ai	184	3	digital	digital	JJ
07-kim-ai	184	4	assistants	assistant	NNS
07-kim-ai	184	5	become	become	VBP
07-kim-ai	184	6	more	more	RBR
07-kim-ai	184	7	capable	capable	JJ
07-kim-ai	184	8	,	,	,
07-kim-ai	184	9	they	-PRON-	PRP
07-kim-ai	184	10	will	will	MD
07-kim-ai	184	11	go	go	VB
07-kim-ai	184	12	beyond	beyond	IN
07-kim-ai	184	13	listing	list	VBG
07-kim-ai	184	14	available	available	JJ
07-kim-ai	184	15	information	information	NN
07-kim-ai	184	16	.	.	.
07-kim-ai	185	1	They	-PRON-	PRP
07-kim-ai	185	2	will	will	MD
07-kim-ai	185	3	further	further	RB
07-kim-ai	185	4	choose	choose	VB
07-kim-ai	185	5	what	what	WP
07-kim-ai	185	6	they	-PRON-	PRP
07-kim-ai	185	7	deem	deem	VBP
07-kim-ai	185	8	to	to	TO
07-kim-ai	185	9	be	be	VB
07-kim-ai	185	10	most	most	RBS
07-kim-ai	185	11	relevant	relevant	JJ
07-kim-ai	185	12	to	to	IN
07-kim-ai	185	13	users	user	NNS
07-kim-ai	185	14	and	and	CC
07-kim-ai	185	15	proceed	proceed	VB
07-kim-ai	185	16	to	to	TO
07-kim-ai	185	17	recommend	recommend	VB
07-kim-ai	185	18	or	or	CC
07-kim-ai	185	19	autonomously	autonomously	RB
07-kim-ai	185	20	execute	execute	VB
07-kim-ai	185	21	the	the	DT
07-kim-ai	185	22	best	good	JJS
07-kim-ai	185	23	course	course	NN
07-kim-ai	185	24	of	of	IN
07-kim-ai	185	25	action.22	action.22	NNP
07-kim-ai	185	26	Other	Other	NNP
07-kim-ai	185	27	AI	AI	NNP
07-kim-ai	185	28	-	-	HYPH
07-kim-ai	185	29	driven	drive	VBN
07-kim-ai	185	30	features	feature	NNS
07-kim-ai	185	31	,	,	,
07-kim-ai	185	32	such	such	JJ
07-kim-ai	185	33	as	as	IN
07-kim-ai	185	34	extracting	extract	VBG
07-kim-ai	185	35	key	key	JJ
07-kim-ai	185	36	information	information	NN
07-kim-ai	185	37	or	or	CC
07-kim-ai	185	38	generat-	generat-	NN
07-kim-ai	185	39	ing	e	VBG
07-kim-ai	185	40	a	a	DT
07-kim-ai	185	41	summary	summary	NN
07-kim-ai	185	42	of	of	IN
07-kim-ai	185	43	a	a	DT
07-kim-ai	185	44	large	large	JJ
07-kim-ai	185	45	amount	amount	NN
07-kim-ai	185	46	of	of	IN
07-kim-ai	185	47	information	information	NN
07-kim-ai	185	48	,	,	,
07-kim-ai	185	49	are	be	VBP
07-kim-ai	185	50	also	also	RB
07-kim-ai	185	51	likely	likely	JJ
07-kim-ai	185	52	to	to	TO
07-kim-ai	185	53	be	be	VB
07-kim-ai	185	54	included	include	VBN
07-kim-ai	185	55	in	in	IN
07-kim-ai	185	56	future	future	JJ
07-kim-ai	185	57	informa-	informa-	JJ
07-kim-ai	185	58	tion	tion	NN
07-kim-ai	185	59	systems	system	NNS
07-kim-ai	185	60	,	,	,
07-kim-ai	185	61	and	and	CC
07-kim-ai	185	62	they	-PRON-	PRP
07-kim-ai	185	63	may	may	MD
07-kim-ai	185	64	deliver	deliver	VB
07-kim-ai	185	65	key	key	JJ
07-kim-ai	185	66	information	information	NN
07-kim-ai	185	67	or	or	CC
07-kim-ai	185	68	summaries	summary	NNS
07-kim-ai	185	69	even	even	RB
07-kim-ai	185	70	before	before	IN
07-kim-ai	185	71	the	the	DT
07-kim-ai	185	72	request	request	NN
07-kim-ai	185	73	is	be	VBZ
07-kim-ai	185	74	made	make	VBN
07-kim-ai	185	75	based	base	VBN
07-kim-ai	185	76	upon	upon	IN
07-kim-ai	185	77	constant	constant	JJ
07-kim-ai	185	78	monitoring	monitoring	NN
07-kim-ai	185	79	of	of	IN
07-kim-ai	185	80	the	the	DT
07-kim-ai	185	81	user	user	NN
07-kim-ai	185	82	’s	’s	POS
07-kim-ai	185	83	activities	activity	NNS
07-kim-ai	185	84	.	.	.
07-kim-ai	186	1	In	in	IN
07-kim-ai	186	2	such	such	PDT
07-kim-ai	186	3	a	a	DT
07-kim-ai	186	4	scenario	scenario	NN
07-kim-ai	186	5	,	,	,
07-kim-ai	186	6	an	an	DT
07-kim-ai	186	7	information	information	NN
07-kim-ai	186	8	seeker	seeker	NN
07-kim-ai	186	9	’s	’s	POS
07-kim-ai	186	10	cognitive	cognitive	JJ
07-kim-ai	186	11	agency	agency	NN
07-kim-ai	186	12	is	be	VBZ
07-kim-ai	186	13	likely	likely	RB
07-kim-ai	186	14	be	be	VB
07-kim-ai	186	15	undermined	undermine	VBN
07-kim-ai	186	16	.	.	.
07-kim-ai	187	1	Cru-	Cru-	NNP
07-kim-ai	187	2	cial	cial	JJ
07-kim-ai	187	3	to	to	IN
07-kim-ai	187	4	cognitive	cognitive	JJ
07-kim-ai	187	5	agency	agency	NN
07-kim-ai	187	6	is	be	VBZ
07-kim-ai	187	7	the	the	DT
07-kim-ai	187	8	mental	mental	JJ
07-kim-ai	187	9	capacity	capacity	NN
07-kim-ai	187	10	to	to	TO
07-kim-ai	187	11	critically	critically	RB
07-kim-ai	187	12	review	review	VB
07-kim-ai	187	13	a	a	DT
07-kim-ai	187	14	variety	variety	NN
07-kim-ai	187	15	of	of	IN
07-kim-ai	187	16	information	information	NN
07-kim-ai	187	17	,	,	,
07-kim-ai	187	18	judge	judge	VBP
07-kim-ai	187	19	what	what	WP
07-kim-ai	187	20	is	be	VBZ
07-kim-ai	187	21	and	and	CC
07-kim-ai	187	22	is	be	VBZ
07-kim-ai	187	23	not	not	RB
07-kim-ai	187	24	relevant	relevant	JJ
07-kim-ai	187	25	,	,	,
07-kim-ai	187	26	and	and	CC
07-kim-ai	187	27	interpret	interpret	VB
07-kim-ai	187	28	how	how	WRB
07-kim-ai	187	29	they	-PRON-	PRP
07-kim-ai	187	30	relate	relate	VBP
07-kim-ai	187	31	to	to	IN
07-kim-ai	187	32	other	other	JJ
07-kim-ai	187	33	existing	exist	VBG
07-kim-ai	187	34	beliefs	belief	NNS
07-kim-ai	187	35	and	and	CC
07-kim-ai	187	36	opinions	opinion	NNS
07-kim-ai	187	37	.	.	.
07-kim-ai	188	1	If	if	IN
07-kim-ai	188	2	AI	AI	NNP
07-kim-ai	188	3	assumes	assume	VBZ
07-kim-ai	188	4	those	those	DT
07-kim-ai	188	5	tasks	task	NNS
07-kim-ai	188	6	,	,	,
07-kim-ai	188	7	the	the	DT
07-kim-ai	188	8	opportunities	opportunity	NNS
07-kim-ai	188	9	for	for	IN
07-kim-ai	188	10	information	information	NN
07-kim-ai	188	11	seekers	seeker	NNS
07-kim-ai	188	12	to	to	TO
07-kim-ai	188	13	exercise	exercise	VB
07-kim-ai	188	14	their	-PRON-	PRP$
07-kim-ai	188	15	own	own	JJ
07-kim-ai	188	16	cogni-	cogni-	NN
07-kim-ai	188	17	tive	tive	JJ
07-kim-ai	188	18	agency	agency	NN
07-kim-ai	188	19	will	will	MD
07-kim-ai	188	20	surely	surely	RB
07-kim-ai	188	21	decrease	decrease	VB
07-kim-ai	188	22	.	.	.
07-kim-ai	189	1	Cognitive	cognitive	JJ
07-kim-ai	189	2	deskilling	deskilling	NN
07-kim-ai	189	3	and	and	CC
07-kim-ai	189	4	the	the	DT
07-kim-ai	189	5	subsequent	subsequent	JJ
07-kim-ai	189	6	weakening	weakening	NN
07-kim-ai	189	7	of	of	IN
07-kim-ai	189	8	people	people	NNS
07-kim-ai	189	9	’s	’s	POS
07-kim-ai	189	10	agency	agency	NN
07-kim-ai	189	11	in	in	IN
07-kim-ai	189	12	the	the	DT
07-kim-ai	189	13	AI	AI	NNP
07-kim-ai	189	14	-powered	-powere	VBN
07-kim-ai	189	15	automated	automate	VBN
07-kim-ai	189	16	information	information	NN
07-kim-ai	189	17	environment	environment	NN
07-kim-ai	189	18	presents	present	VBZ
07-kim-ai	189	19	an	an	DT
07-kim-ai	189	20	ethical	ethical	JJ
07-kim-ai	189	21	challenge	challenge	NN
07-kim-ai	189	22	because	because	IN
07-kim-ai	189	23	such	such	JJ
07-kim-ai	189	24	agency	agency	NN
07-kim-ai	189	25	is	be	VBZ
07-kim-ai	189	26	necessary	necessary	JJ
07-kim-ai	189	27	for	for	IN
07-kim-ai	189	28	a	a	DT
07-kim-ai	189	29	person	person	NN
07-kim-ai	189	30	to	to	TO
07-kim-ai	189	31	be	be	VB
07-kim-ai	189	32	a	a	DT
07-kim-ai	189	33	fully	fully	RB
07-kim-ai	189	34	functioning	function	VBG
07-kim-ai	189	35	moral	moral	JJ
07-kim-ai	189	36	agent	agent	NN
07-kim-ai	189	37	in	in	IN
07-kim-ai	189	38	society.23	society.23	NNP
07-kim-ai	189	39	(	(	-LRB-
07-kim-ai	189	40	iii	iii	NNP
07-kim-ai	189	41	)	)	-RRB-
07-kim-ai	189	42	Social	social	JJ
07-kim-ai	189	43	impact	impact	NN
07-kim-ai	189	44	of	of	IN
07-kim-ai	189	45	scholarship	scholarship	NN
07-kim-ai	189	46	and	and	CC
07-kim-ai	189	47	research	research	NN
07-kim-ai	189	48	from	from	IN
07-kim-ai	189	49	flawed	flawed	JJ
07-kim-ai	189	50	AI	AI	NNP
07-kim-ai	189	51	algorithms	algorithm	NNS
07-kim-ai	189	52	Previously	previously	RB
07-kim-ai	189	53	,	,	,
07-kim-ai	189	54	we	-PRON-	PRP
07-kim-ai	189	55	have	have	VBP
07-kim-ai	189	56	seen	see	VBN
07-kim-ai	189	57	that	that	IN
07-kim-ai	189	58	deep	deep	JJ
07-kim-ai	189	59	learning	learning	NN
07-kim-ai	189	60	applications	application	NNS
07-kim-ai	189	61	are	be	VBP
07-kim-ai	189	62	opaque	opaque	JJ
07-kim-ai	189	63	to	to	IN
07-kim-ai	189	64	human	human	JJ
07-kim-ai	189	65	understanding	understanding	NN
07-kim-ai	189	66	.	.	.
07-kim-ai	190	1	This	this	DT
07-kim-ai	190	2	lack	lack	NN
07-kim-ai	190	3	of	of	IN
07-kim-ai	190	4	transparency	transparency	NN
07-kim-ai	190	5	and	and	CC
07-kim-ai	190	6	explainability	explainability	NN
07-kim-ai	190	7	raises	raise	VBZ
07-kim-ai	190	8	a	a	DT
07-kim-ai	190	9	question	question	NN
07-kim-ai	190	10	of	of	IN
07-kim-ai	190	11	whether	whether	IN
07-kim-ai	190	12	it	-PRON-	PRP
07-kim-ai	190	13	is	be	VBZ
07-kim-ai	190	14	moral	moral	JJ
07-kim-ai	190	15	to	to	TO
07-kim-ai	190	16	rely	rely	VB
07-kim-ai	190	17	on	on	IN
07-kim-ai	190	18	AI	AI	NNP
07-kim-ai	190	19	-	-	HYPH
07-kim-ai	190	20	powered	power	VBN
07-kim-ai	190	21	military	military	JJ
07-kim-ai	190	22	robots	robot	NNS
07-kim-ai	190	23	for	for	IN
07-kim-ai	190	24	life	life	NN
07-kim-ai	190	25	-	-	HYPH
07-kim-ai	190	26	or	or	CC
07-kim-ai	190	27	-	-	HYPH
07-kim-ai	190	28	death	death	NN
07-kim-ai	190	29	decisions	decision	NNS
07-kim-ai	190	30	.	.	.
07-kim-ai	191	1	Does	do	VBZ
07-kim-ai	191	2	the	the	DT
07-kim-ai	191	3	AI	ai	NN
07-kim-ai	191	4	-	-	HYPH
07-kim-ai	191	5	powered	power	VBN
07-kim-ai	191	6	information	information	NN
07-kim-ai	191	7	envi-	envi-	IN
07-kim-ai	191	8	ronment	ronment	NN
07-kim-ai	191	9	have	have	VBP
07-kim-ai	191	10	a	a	DT
07-kim-ai	191	11	similar	similar	JJ
07-kim-ai	191	12	problem	problem	NN
07-kim-ai	191	13	?	?	.
07-kim-ai	192	1	Machine	machine	NN
07-kim-ai	192	2	learning	learning	NN
07-kim-ai	192	3	applications	application	NNS
07-kim-ai	192	4	base	base	VBP
07-kim-ai	192	5	their	-PRON-	PRP$
07-kim-ai	192	6	recommendations	recommendation	NNS
07-kim-ai	192	7	and	and	CC
07-kim-ai	192	8	predictions	prediction	NNS
07-kim-ai	192	9	upon	upon	IN
07-kim-ai	192	10	the	the	DT
07-kim-ai	192	11	pat-	pat-	NN
07-kim-ai	192	12	terns	tern	NNS
07-kim-ai	192	13	in	in	IN
07-kim-ai	192	14	past	past	JJ
07-kim-ai	192	15	data	datum	NNS
07-kim-ai	192	16	.	.	.
07-kim-ai	193	1	Their	-PRON-	PRP$
07-kim-ai	193	2	predictions	prediction	NNS
07-kim-ai	193	3	and	and	CC
07-kim-ai	193	4	recommendations	recommendation	NNS
07-kim-ai	193	5	are	be	VBP
07-kim-ai	193	6	in	in	IN
07-kim-ai	193	7	this	this	DT
07-kim-ai	193	8	sense	sense	NN
07-kim-ai	193	9	inherently	inherently	RB
07-kim-ai	193	10	conser-	conser-	IN
07-kim-ai	193	11	vative	vative	JJ
07-kim-ai	193	12	.	.	.
07-kim-ai	194	1	They	-PRON-	PRP
07-kim-ai	194	2	also	also	RB
07-kim-ai	194	3	become	become	VBP
07-kim-ai	194	4	outdated	outdated	JJ
07-kim-ai	194	5	when	when	WRB
07-kim-ai	194	6	they	-PRON-	PRP
07-kim-ai	194	7	fail	fail	VBP
07-kim-ai	194	8	to	to	TO
07-kim-ai	194	9	reflect	reflect	VB
07-kim-ai	194	10	new	new	JJ
07-kim-ai	194	11	social	social	JJ
07-kim-ai	194	12	views	view	NNS
07-kim-ai	194	13	and	and	CC
07-kim-ai	194	14	material	material	NN
07-kim-ai	194	15	con-	con-	NN
07-kim-ai	194	16	ditions	dition	NNS
07-kim-ai	194	17	that	that	WDT
07-kim-ai	194	18	no	no	RB
07-kim-ai	194	19	longer	long	RBR
07-kim-ai	194	20	fit	fit	VBP
07-kim-ai	194	21	the	the	DT
07-kim-ai	194	22	past	past	JJ
07-kim-ai	194	23	patterns	pattern	NNS
07-kim-ai	194	24	.	.	.
07-kim-ai	195	1	Furthermore	furthermore	RB
07-kim-ai	195	2	,	,	,
07-kim-ai	195	3	each	each	DT
07-kim-ai	195	4	data	data	NN
07-kim-ai	195	5	set	set	NN
07-kim-ai	195	6	is	be	VBZ
07-kim-ai	195	7	a	a	DT
07-kim-ai	195	8	social	social	JJ
07-kim-ai	195	9	construct	construct	NN
07-kim-ai	195	10	that	that	WDT
07-kim-ai	195	11	reflects	reflect	VBZ
07-kim-ai	195	12	particular	particular	JJ
07-kim-ai	195	13	values	value	NNS
07-kim-ai	195	14	and	and	CC
07-kim-ai	195	15	choices	choice	NNS
07-kim-ai	195	16	such	such	JJ
07-kim-ai	195	17	as	as	IN
07-kim-ai	195	18	who	who	WP
07-kim-ai	195	19	decided	decide	VBD
07-kim-ai	195	20	to	to	TO
07-kim-ai	195	21	collect	collect	VB
07-kim-ai	195	22	the	the	DT
07-kim-ai	195	23	data	datum	NNS
07-kim-ai	195	24	and	and	CC
07-kim-ai	195	25	for	for	IN
07-kim-ai	195	26	what	what	WP
07-kim-ai	195	27	pur-	pur-	NN
07-kim-ai	195	28	pose	pose	VBP
07-kim-ai	195	29	;	;	:
07-kim-ai	195	30	who	who	WP
07-kim-ai	195	31	labeled	label	VBD
07-kim-ai	195	32	data	datum	NNS
07-kim-ai	195	33	;	;	:
07-kim-ai	195	34	what	what	WDT
07-kim-ai	195	35	criteria	criterion	NNS
07-kim-ai	195	36	or	or	CC
07-kim-ai	195	37	beliefs	belief	NNS
07-kim-ai	195	38	guided	guide	VBD
07-kim-ai	195	39	such	such	JJ
07-kim-ai	195	40	labeling	labeling	NN
07-kim-ai	195	41	;	;	:
07-kim-ai	195	42	what	what	WP
07-kim-ai	195	43	taxonomies	taxonomy	NNS
07-kim-ai	195	44	were	be	VBD
07-kim-ai	195	45	used	use	VBN
07-kim-ai	195	46	and	and	CC
07-kim-ai	195	47	why	why	WRB
07-kim-ai	195	48	(	(	-LRB-
07-kim-ai	195	49	Davis	Davis	NNP
07-kim-ai	195	50	2020	2020	CD
07-kim-ai	195	51	)	)	-RRB-
07-kim-ai	195	52	.	.	.
07-kim-ai	196	1	No	no	DT
07-kim-ai	196	2	data	data	NN
07-kim-ai	196	3	set	set	NN
07-kim-ai	196	4	can	can	MD
07-kim-ai	196	5	capture	capture	VB
07-kim-ai	196	6	all	all	DT
07-kim-ai	196	7	variables	variable	NNS
07-kim-ai	196	8	and	and	CC
07-kim-ai	196	9	elements	element	NNS
07-kim-ai	196	10	of	of	IN
07-kim-ai	196	11	the	the	DT
07-kim-ai	196	12	phenomenon	phenomenon	NN
07-kim-ai	196	13	that	that	WDT
07-kim-ai	196	14	it	-PRON-	PRP
07-kim-ai	196	15	describes	describe	VBZ
07-kim-ai	196	16	.	.	.
07-kim-ai	197	1	Furthermore	furthermore	RB
07-kim-ai	197	2	,	,	,
07-kim-ai	197	3	data	data	NN
07-kim-ai	197	4	sets	set	NNS
07-kim-ai	197	5	used	use	VBN
07-kim-ai	197	6	for	for	IN
07-kim-ai	197	7	training	train	VBG
07-kim-ai	197	8	machine	machine	NN
07-kim-ai	197	9	learning	learning	NN
07-kim-ai	197	10	and	and	CC
07-kim-ai	197	11	deep	deep	JJ
07-kim-ai	197	12	learning	learning	NN
07-kim-ai	197	13	algorithms	algorithm	NNS
07-kim-ai	197	14	may	may	MD
07-kim-ai	197	15	not	not	RB
07-kim-ai	197	16	be	be	VB
07-kim-ai	197	17	representational	representational	JJ
07-kim-ai	197	18	samples	sample	NNS
07-kim-ai	197	19	for	for	IN
07-kim-ai	197	20	all	all	DT
07-kim-ai	197	21	relevant	relevant	JJ
07-kim-ai	197	22	subgroups	subgroup	NNS
07-kim-ai	197	23	.	.	.
07-kim-ai	198	1	In	in	IN
07-kim-ai	198	2	such	such	PDT
07-kim-ai	198	3	a	a	DT
07-kim-ai	198	4	case	case	NN
07-kim-ai	198	5	,	,	,
07-kim-ai	198	6	an	an	DT
07-kim-ai	198	7	al-	al-	JJ
07-kim-ai	198	8	gorithm	gorithm	NN
07-kim-ai	198	9	trained	train	VBN
07-kim-ai	198	10	by	by	IN
07-kim-ai	198	11	such	such	PDT
07-kim-ai	198	12	a	a	DT
07-kim-ai	198	13	data	data	NN
07-kim-ai	198	14	set	set	NN
07-kim-ai	198	15	will	will	MD
07-kim-ai	198	16	produce	produce	VB
07-kim-ai	198	17	skewed	skewed	JJ
07-kim-ai	198	18	results	result	NNS
07-kim-ai	198	19	.	.	.
07-kim-ai	199	1	Creating	create	VBG
07-kim-ai	199	2	a	a	DT
07-kim-ai	199	3	large	large	JJ
07-kim-ai	199	4	data	data	NN
07-kim-ai	199	5	set	set	NN
07-kim-ai	199	6	is	be	VBZ
07-kim-ai	199	7	also	also	RB
07-kim-ai	199	8	costly	costly	JJ
07-kim-ai	199	9	.	.	.
07-kim-ai	200	1	Consequently	consequently	RB
07-kim-ai	200	2	,	,	,
07-kim-ai	200	3	developers	developer	NNS
07-kim-ai	200	4	often	often	RB
07-kim-ai	200	5	simply	simply	RB
07-kim-ai	200	6	take	take	VBP
07-kim-ai	200	7	the	the	DT
07-kim-ai	200	8	data	data	NN
07-kim-ai	200	9	sets	set	NNS
07-kim-ai	200	10	available	available	JJ
07-kim-ai	200	11	to	to	IN
07-kim-ai	200	12	them	-PRON-	PRP
07-kim-ai	200	13	.	.	.
07-kim-ai	201	1	Those	those	DT
07-kim-ai	201	2	data	data	NN
07-kim-ai	201	3	sets	set	NNS
07-kim-ai	201	4	are	be	VBP
07-kim-ai	201	5	likely	likely	JJ
07-kim-ai	201	6	to	to	TO
07-kim-ai	201	7	come	come	VB
07-kim-ai	201	8	with	with	IN
07-kim-ai	201	9	inherent	inherent	JJ
07-kim-ai	201	10	limitations	limitation	NNS
07-kim-ai	201	11	such	such	JJ
07-kim-ai	201	12	as	as	IN
07-kim-ai	201	13	omissions	omission	NNS
07-kim-ai	201	14	,	,	,
07-kim-ai	201	15	inaccuracies	inaccuracy	NNS
07-kim-ai	201	16	,	,	,
07-kim-ai	201	17	errors	error	NNS
07-kim-ai	201	18	,	,	,
07-kim-ai	201	19	and	and	CC
07-kim-ai	201	20	hidden	hide	VBN
07-kim-ai	201	21	biases	bias	NNS
07-kim-ai	201	22	.	.	.
07-kim-ai	202	1	22Needless	22needless	CD
07-kim-ai	202	2	to	to	TO
07-kim-ai	202	3	say	say	VB
07-kim-ai	202	4	,	,	,
07-kim-ai	202	5	this	this	DT
07-kim-ai	202	6	is	be	VBZ
07-kim-ai	202	7	a	a	DT
07-kim-ai	202	8	highly	highly	RB
07-kim-ai	202	9	simplified	simplify	VBN
07-kim-ai	202	10	scenario	scenario	NN
07-kim-ai	202	11	.	.	.
07-kim-ai	203	1	Those	those	DT
07-kim-ai	203	2	features	feature	NNS
07-kim-ai	203	3	can	can	MD
07-kim-ai	203	4	also	also	RB
07-kim-ai	203	5	be	be	VB
07-kim-ai	203	6	built	build	VBN
07-kim-ai	203	7	in	in	IN
07-kim-ai	203	8	the	the	DT
07-kim-ai	203	9	information	information	NN
07-kim-ai	203	10	system	system	NN
07-kim-ai	203	11	itself	-PRON-	PRP
07-kim-ai	203	12	rather	rather	RB
07-kim-ai	203	13	than	than	IN
07-kim-ai	203	14	being	be	VBG
07-kim-ai	203	15	delivered	deliver	VBN
07-kim-ai	203	16	by	by	IN
07-kim-ai	203	17	a	a	DT
07-kim-ai	203	18	digital	digital	JJ
07-kim-ai	203	19	assistant	assistant	NN
07-kim-ai	203	20	.	.	.
07-kim-ai	204	1	23Outside	23outside	CD
07-kim-ai	204	2	of	of	IN
07-kim-ai	204	3	the	the	DT
07-kim-ai	204	4	automated	automated	JJ
07-kim-ai	204	5	information	information	NN
07-kim-ai	204	6	environment	environment	NN
07-kim-ai	204	7	,	,	,
07-kim-ai	204	8	AI	AI	NNP
07-kim-ai	204	9	has	have	VBZ
07-kim-ai	204	10	a	a	DT
07-kim-ai	204	11	strong	strong	JJ
07-kim-ai	204	12	potential	potential	NN
07-kim-ai	204	13	to	to	TO
07-kim-ai	204	14	engender	engender	VB
07-kim-ai	204	15	moral	moral	JJ
07-kim-ai	204	16	deskilling	deskilling	NN
07-kim-ai	204	17	.	.	.
07-kim-ai	205	1	Vallor	Vallor	NNP
07-kim-ai	205	2	(	(	-LRB-
07-kim-ai	205	3	2015	2015	CD
07-kim-ai	205	4	)	)	-RRB-
07-kim-ai	205	5	points	point	VBZ
07-kim-ai	205	6	out	out	RP
07-kim-ai	205	7	that	that	IN
07-kim-ai	205	8	automated	automate	VBN
07-kim-ai	205	9	weapons	weapon	NNS
07-kim-ai	205	10	will	will	MD
07-kim-ai	205	11	lead	lead	VB
07-kim-ai	205	12	to	to	IN
07-kim-ai	205	13	soldiers	soldier	NNS
07-kim-ai	205	14	’	’	POS
07-kim-ai	205	15	moral	moral	JJ
07-kim-ai	205	16	deskilling	deskilling	NN
07-kim-ai	205	17	in	in	IN
07-kim-ai	205	18	the	the	DT
07-kim-ai	205	19	use	use	NN
07-kim-ai	205	20	of	of	IN
07-kim-ai	205	21	military	military	JJ
07-kim-ai	205	22	force	force	NN
07-kim-ai	205	23	;	;	:
07-kim-ai	205	24	new	new	NNP
07-kim-ai	205	25	me-	me-	XX
07-kim-ai	205	26	dia	dia	VB
07-kim-ai	205	27	practices	practice	NNS
07-kim-ai	205	28	of	of	IN
07-kim-ai	205	29	multitasking	multitasking	NN
07-kim-ai	205	30	may	may	MD
07-kim-ai	205	31	result	result	VB
07-kim-ai	205	32	in	in	IN
07-kim-ai	205	33	deskilling	deskille	VBG
07-kim-ai	205	34	in	in	IN
07-kim-ai	205	35	moral	moral	JJ
07-kim-ai	205	36	attention	attention	NN
07-kim-ai	205	37	;	;	:
07-kim-ai	205	38	and	and	CC
07-kim-ai	205	39	social	social	JJ
07-kim-ai	205	40	robots	robot	NNS
07-kim-ai	205	41	can	can	MD
07-kim-ai	205	42	cause	cause	VB
07-kim-ai	205	43	moral	moral	JJ
07-kim-ai	205	44	deskilling	deskilling	NN
07-kim-ai	205	45	in	in	IN
07-kim-ai	205	46	practices	practice	NNS
07-kim-ai	205	47	of	of	IN
07-kim-ai	205	48	human	human	JJ
07-kim-ai	205	49	caregiving	caregiving	NN
07-kim-ai	205	50	.	.	.
07-kim-ai	206	1	Kim	Kim	NNP
07-kim-ai	206	2	81	81	CD
07-kim-ai	206	3	AI	AI	NNP
07-kim-ai	206	4	algorithms	algorithm	NNS
07-kim-ai	206	5	trained	train	VBD
07-kim-ai	206	6	with	with	IN
07-kim-ai	206	7	these	these	DT
07-kim-ai	206	8	flawed	flawed	JJ
07-kim-ai	206	9	data	data	NN
07-kim-ai	206	10	sets	set	NNS
07-kim-ai	206	11	can	can	MD
07-kim-ai	206	12	fail	fail	VB
07-kim-ai	206	13	unexpectedly	unexpectedly	RB
07-kim-ai	206	14	,	,	,
07-kim-ai	206	15	revealing	reveal	VBG
07-kim-ai	206	16	those	those	DT
07-kim-ai	206	17	limi-	limi-	JJ
07-kim-ai	206	18	tations	tation	NNS
07-kim-ai	206	19	.	.	.
07-kim-ai	207	1	For	for	IN
07-kim-ai	207	2	example	example	NN
07-kim-ai	207	3	,	,	,
07-kim-ai	207	4	it	-PRON-	PRP
07-kim-ai	207	5	has	have	VBZ
07-kim-ai	207	6	been	be	VBN
07-kim-ai	207	7	reported	report	VBN
07-kim-ai	207	8	that	that	IN
07-kim-ai	207	9	the	the	DT
07-kim-ai	207	10	success	success	NN
07-kim-ai	207	11	rate	rate	NN
07-kim-ai	207	12	of	of	IN
07-kim-ai	207	13	a	a	DT
07-kim-ai	207	14	facial	facial	JJ
07-kim-ai	207	15	recognition	recognition	NN
07-kim-ai	207	16	algorithm	algorithm	NN
07-kim-ai	207	17	plunges	plunge	NNS
07-kim-ai	207	18	from	from	IN
07-kim-ai	207	19	99	99	CD
07-kim-ai	207	20	%	%	NN
07-kim-ai	207	21	to	to	TO
07-kim-ai	207	22	35	35	CD
07-kim-ai	207	23	%	%	NN
07-kim-ai	207	24	when	when	WRB
07-kim-ai	207	25	the	the	DT
07-kim-ai	207	26	group	group	NN
07-kim-ai	207	27	of	of	IN
07-kim-ai	207	28	subjects	subject	NNS
07-kim-ai	207	29	changes	change	NNS
07-kim-ai	207	30	from	from	IN
07-kim-ai	207	31	white	white	JJ
07-kim-ai	207	32	men	man	NNS
07-kim-ai	207	33	to	to	IN
07-kim-ai	207	34	dark	dark	JJ
07-kim-ai	207	35	-	-	HYPH
07-kim-ai	207	36	skinned	skinned	JJ
07-kim-ai	207	37	women	woman	NNS
07-kim-ai	207	38	because	because	IN
07-kim-ai	207	39	it	-PRON-	PRP
07-kim-ai	207	40	was	be	VBD
07-kim-ai	207	41	trained	train	VBN
07-kim-ai	207	42	mostly	mostly	RB
07-kim-ai	207	43	with	with	IN
07-kim-ai	207	44	the	the	DT
07-kim-ai	207	45	photographs	photograph	NNS
07-kim-ai	207	46	of	of	IN
07-kim-ai	207	47	white	white	JJ
07-kim-ai	207	48	men	man	NNS
07-kim-ai	207	49	(	(	-LRB-
07-kim-ai	207	50	Lohr	Lohr	NNP
07-kim-ai	207	51	2018	2018	CD
07-kim-ai	207	52	)	)	-RRB-
07-kim-ai	207	53	.	.	.
07-kim-ai	208	1	Adopt-	Adopt-	NNP
07-kim-ai	208	2	ing	e	VBG
07-kim-ai	208	3	such	such	PDT
07-kim-ai	208	4	a	a	DT
07-kim-ai	208	5	faulty	faulty	JJ
07-kim-ai	208	6	algorithm	algorithm	NN
07-kim-ai	208	7	for	for	IN
07-kim-ai	208	8	any	any	DT
07-kim-ai	208	9	real	real	JJ
07-kim-ai	208	10	-	-	HYPH
07-kim-ai	208	11	life	life	NN
07-kim-ai	208	12	use	use	NN
07-kim-ai	208	13	at	at	IN
07-kim-ai	208	14	a	a	DT
07-kim-ai	208	15	large	large	JJ
07-kim-ai	208	16	scale	scale	NN
07-kim-ai	208	17	would	would	MD
07-kim-ai	208	18	be	be	VB
07-kim-ai	208	19	entirely	entirely	RB
07-kim-ai	208	20	unethical	unethical	JJ
07-kim-ai	208	21	.	.	.
07-kim-ai	209	1	For	for	IN
07-kim-ai	209	2	the	the	DT
07-kim-ai	209	3	context	context	NN
07-kim-ai	209	4	of	of	IN
07-kim-ai	209	5	libraries	library	NNS
07-kim-ai	209	6	,	,	,
07-kim-ai	209	7	imagine	imagine	VB
07-kim-ai	209	8	using	use	VBG
07-kim-ai	209	9	such	such	PDT
07-kim-ai	209	10	a	a	DT
07-kim-ai	209	11	face	face	NN
07-kim-ai	209	12	-	-	HYPH
07-kim-ai	209	13	recognition	recognition	NN
07-kim-ai	209	14	algorithm	algorithm	NN
07-kim-ai	209	15	to	to	TO
07-kim-ai	209	16	generate	generate	VB
07-kim-ai	209	17	metadata	metadata	NN
07-kim-ai	209	18	for	for	IN
07-kim-ai	209	19	digitized	digitized	JJ
07-kim-ai	209	20	historical	historical	JJ
07-kim-ai	209	21	photographs	photograph	NNS
07-kim-ai	209	22	or	or	CC
07-kim-ai	209	23	a	a	DT
07-kim-ai	209	24	similarly	similarly	RB
07-kim-ai	209	25	flawed	flawed	JJ
07-kim-ai	209	26	audio	audio	JJ
07-kim-ai	209	27	transcription	transcription	NN
07-kim-ai	209	28	algorithm	algorithm	NNP
07-kim-ai	209	29	to	to	TO
07-kim-ai	209	30	transcribe	transcribe	VB
07-kim-ai	209	31	archival	archival	NNP
07-kim-ai	209	32	audio	audio	NNP
07-kim-ai	209	33	recordings	recording	NNS
07-kim-ai	209	34	.	.	.
07-kim-ai	210	1	Just	just	RB
07-kim-ai	210	2	like	like	IN
07-kim-ai	210	3	those	those	DT
07-kim-ai	210	4	faulty	faulty	JJ
07-kim-ai	210	5	algorithms	algorithm	NNS
07-kim-ai	210	6	,	,	,
07-kim-ai	210	7	an	an	DT
07-kim-ai	210	8	AI	AI	NNP
07-kim-ai	210	9	-	-	HYPH
07-kim-ai	210	10	powered	power	VBN
07-kim-ai	210	11	automated	automated	JJ
07-kim-ai	210	12	information	information	NN
07-kim-ai	210	13	environment	environment	NN
07-kim-ai	210	14	can	can	MD
07-kim-ai	210	15	produce	produce	VB
07-kim-ai	210	16	information	information	NN
07-kim-ai	210	17	,	,	,
07-kim-ai	210	18	recommendations	recommendation	NNS
07-kim-ai	210	19	,	,	,
07-kim-ai	210	20	and	and	CC
07-kim-ai	210	21	predictions	prediction	NNS
07-kim-ai	210	22	affected	affect	VBN
07-kim-ai	210	23	by	by	IN
07-kim-ai	210	24	similar	similar	JJ
07-kim-ai	210	25	limitations	limitation	NNS
07-kim-ai	210	26	existing	exist	VBG
07-kim-ai	210	27	in	in	IN
07-kim-ai	210	28	many	many	JJ
07-kim-ai	210	29	data	datum	NNS
07-kim-ai	210	30	sets	set	NNS
07-kim-ai	210	31	.	.	.
07-kim-ai	211	1	The	the	DT
07-kim-ai	211	2	more	more	RBR
07-kim-ai	211	3	seamless	seamless	NN
07-kim-ai	211	4	such	such	PDT
07-kim-ai	211	5	an	an	DT
07-kim-ai	211	6	information	information	NN
07-kim-ai	211	7	environment	environment	NN
07-kim-ai	211	8	is	be	VBZ
07-kim-ai	211	9	,	,	,
07-kim-ai	211	10	the	the	DT
07-kim-ai	211	11	more	more	RBR
07-kim-ai	211	12	invisible	invisible	JJ
07-kim-ai	211	13	those	those	DT
07-kim-ai	211	14	limitations	limitation	NNS
07-kim-ai	211	15	become	become	VBP
07-kim-ai	211	16	.	.	.
07-kim-ai	212	1	Automated	automate	VBN
07-kim-ai	212	2	information	information	NN
07-kim-ai	212	3	systems	system	NNS
07-kim-ai	212	4	from	from	IN
07-kim-ai	212	5	libraries	library	NNS
07-kim-ai	212	6	may	may	MD
07-kim-ai	212	7	not	not	RB
07-kim-ai	212	8	be	be	VB
07-kim-ai	212	9	involved	involve	VBN
07-kim-ai	212	10	in	in	IN
07-kim-ai	212	11	decisions	decision	NNS
07-kim-ai	212	12	that	that	WDT
07-kim-ai	212	13	have	have	VBP
07-kim-ai	212	14	a	a	DT
07-kim-ai	212	15	direct	direct	JJ
07-kim-ai	212	16	and	and	CC
07-kim-ai	212	17	immediate	immediate	JJ
07-kim-ai	212	18	impact	impact	NN
07-kim-ai	212	19	on	on	IN
07-kim-ai	212	20	people	people	NNS
07-kim-ai	212	21	’s	’s	POS
07-kim-ai	212	22	lives	life	NNS
07-kim-ai	212	23	,	,	,
07-kim-ai	212	24	such	such	JJ
07-kim-ai	212	25	as	as	IN
07-kim-ai	212	26	setting	set	VBG
07-kim-ai	212	27	a	a	DT
07-kim-ai	212	28	bail	bail	NN
07-kim-ai	212	29	amount	amount	NN
07-kim-ai	212	30	or	or	CC
07-kim-ai	212	31	determining	determine	VBG
07-kim-ai	212	32	the	the	DT
07-kim-ai	212	33	Medicaid	Medicaid	NNP
07-kim-ai	212	34	payment	payment	NN
07-kim-ai	212	35	to	to	TO
07-kim-ai	212	36	be	be	VB
07-kim-ai	212	37	paid.24	paid.24	NNP
07-kim-ai	212	38	But	but	CC
07-kim-ai	212	39	automated	automate	VBD
07-kim-ai	212	40	information	information	NN
07-kim-ai	212	41	systems	system	NNS
07-kim-ai	212	42	that	that	WDT
07-kim-ai	212	43	are	be	VBP
07-kim-ai	212	44	widely	widely	RB
07-kim-ai	212	45	adopted	adopt	VBN
07-kim-ai	212	46	and	and	CC
07-kim-ai	212	47	used	use	VBN
07-kim-ai	212	48	for	for	IN
07-kim-ai	212	49	research	research	NN
07-kim-ai	212	50	and	and	CC
07-kim-ai	212	51	scholarship	scholarship	NN
07-kim-ai	212	52	will	will	MD
07-kim-ai	212	53	impact	impact	VB
07-kim-ai	212	54	real	real	JJ
07-kim-ai	212	55	-	-	HYPH
07-kim-ai	212	56	life	life	NN
07-kim-ai	212	57	policies	policy	NNS
07-kim-ai	212	58	and	and	CC
07-kim-ai	212	59	regulations	regulation	NNS
07-kim-ai	212	60	in	in	IN
07-kim-ai	212	61	areas	area	NNS
07-kim-ai	212	62	such	such	JJ
07-kim-ai	212	63	as	as	IN
07-kim-ai	212	64	healthcare	healthcare	NN
07-kim-ai	212	65	and	and	CC
07-kim-ai	212	66	the	the	DT
07-kim-ai	212	67	economy	economy	NN
07-kim-ai	212	68	.	.	.
07-kim-ai	213	1	Undiscovered	undiscovered	JJ
07-kim-ai	213	2	flaws	flaw	NNS
07-kim-ai	213	3	will	will	MD
07-kim-ai	213	4	undermine	undermine	VB
07-kim-ai	213	5	the	the	DT
07-kim-ai	213	6	validity	validity	NN
07-kim-ai	213	7	of	of	IN
07-kim-ai	213	8	the	the	DT
07-kim-ai	213	9	scholarly	scholarly	JJ
07-kim-ai	213	10	output	output	NN
07-kim-ai	213	11	that	that	WDT
07-kim-ai	213	12	utilized	utilize	VBD
07-kim-ai	213	13	those	those	DT
07-kim-ai	213	14	automated	automate	VBN
07-kim-ai	213	15	information	information	NN
07-kim-ai	213	16	systems	system	NNS
07-kim-ai	213	17	and	and	CC
07-kim-ai	213	18	can	can	MD
07-kim-ai	213	19	further	far	RBR
07-kim-ai	213	20	inflict	inflict	VB
07-kim-ai	213	21	serious	serious	JJ
07-kim-ai	213	22	harm	harm	NN
07-kim-ai	213	23	on	on	IN
07-kim-ai	213	24	certain	certain	JJ
07-kim-ai	213	25	groups	group	NNS
07-kim-ai	213	26	of	of	IN
07-kim-ai	213	27	people	people	NNS
07-kim-ai	213	28	through	through	IN
07-kim-ai	213	29	those	those	DT
07-kim-ai	213	30	policies	policy	NNS
07-kim-ai	213	31	and	and	CC
07-kim-ai	213	32	regulations	regulation	NNS
07-kim-ai	213	33	.	.	.
07-kim-ai	214	1	Moral	Moral	NNP
07-kim-ai	214	2	Intelligence	Intelligence	NNP
07-kim-ai	214	3	and	and	CC
07-kim-ai	214	4	Rethinking	rethink	VBG
07-kim-ai	214	5	the	the	DT
07-kim-ai	214	6	Role	Role	NNP
07-kim-ai	214	7	of	of	IN
07-kim-ai	214	8	AI	AI	NNP
07-kim-ai	214	9	In	in	IN
07-kim-ai	214	10	this	this	DT
07-kim-ai	214	11	chapter	chapter	NN
07-kim-ai	214	12	,	,	,
07-kim-ai	214	13	I	-PRON-	PRP
07-kim-ai	214	14	discussed	discuss	VBD
07-kim-ai	214	15	four	four	CD
07-kim-ai	214	16	significant	significant	JJ
07-kim-ai	214	17	ethical	ethical	JJ
07-kim-ai	214	18	challenges	challenge	NNS
07-kim-ai	214	19	that	that	IN
07-kim-ai	214	20	automating	automate	VBG
07-kim-ai	214	21	decisions	decision	NNS
07-kim-ai	214	22	and	and	CC
07-kim-ai	214	23	ac-	ac-	JJ
07-kim-ai	214	24	tions	tion	NNS
07-kim-ai	214	25	with	with	IN
07-kim-ai	214	26	AI	AI	NNP
07-kim-ai	214	27	presents	present	NNS
07-kim-ai	214	28	:	:	:
07-kim-ai	214	29	(	(	-LRB-
07-kim-ai	214	30	a	a	DT
07-kim-ai	214	31	)	)	-RRB-
07-kim-ai	214	32	moral	moral	JJ
07-kim-ai	214	33	desensitization	desensitization	NN
07-kim-ai	214	34	;	;	,
07-kim-ai	214	35	(	(	-LRB-
07-kim-ai	214	36	b	b	NN
07-kim-ai	214	37	)	)	-RRB-
07-kim-ai	214	38	unintended	unintended	JJ
07-kim-ai	214	39	outcomes	outcome	NNS
07-kim-ai	214	40	;	;	,
07-kim-ai	214	41	(	(	-LRB-
07-kim-ai	214	42	c	c	NN
07-kim-ai	214	43	)	)	-RRB-
07-kim-ai	214	44	surrender	surrender	NN
07-kim-ai	214	45	of	of	IN
07-kim-ai	214	46	moral	moral	JJ
07-kim-ai	214	47	agency	agency	NN
07-kim-ai	214	48	;	;	:
07-kim-ai	214	49	(	(	-LRB-
07-kim-ai	214	50	d	d	NN
07-kim-ai	214	51	)	)	-RRB-
07-kim-ai	214	52	opacity	opacity	NN
07-kim-ai	214	53	in	in	IN
07-kim-ai	214	54	decision	decision	NN
07-kim-ai	214	55	-	-	HYPH
07-kim-ai	214	56	making.25	making.25	NNP
07-kim-ai	214	57	I	-PRON-	PRP
07-kim-ai	214	58	also	also	RB
07-kim-ai	214	59	examined	examine	VBD
07-kim-ai	214	60	somewhat	somewhat	RB
07-kim-ai	214	61	different	different	JJ
07-kim-ai	214	62	but	but	CC
07-kim-ai	214	63	equally	equally	RB
07-kim-ai	214	64	significant	significant	JJ
07-kim-ai	214	65	ethical	ethical	JJ
07-kim-ai	214	66	challenges	challenge	NNS
07-kim-ai	214	67	in	in	IN
07-kim-ai	214	68	relation	relation	NN
07-kim-ai	214	69	to	to	IN
07-kim-ai	214	70	the	the	DT
07-kim-ai	214	71	AI	AI	NNP
07-kim-ai	214	72	-	-	HYPH
07-kim-ai	214	73	powered	power	VBN
07-kim-ai	214	74	automated	automated	JJ
07-kim-ai	214	75	information	information	NN
07-kim-ai	214	76	environment	environment	NN
07-kim-ai	214	77	,	,	,
07-kim-ai	214	78	which	which	WDT
07-kim-ai	214	79	is	be	VBZ
07-kim-ai	214	80	likely	likely	JJ
07-kim-ai	214	81	to	to	TO
07-kim-ai	214	82	surround	surround	VB
07-kim-ai	214	83	us	-PRON-	PRP
07-kim-ai	214	84	in	in	IN
07-kim-ai	214	85	the	the	DT
07-kim-ai	214	86	future	future	NN
07-kim-ai	214	87	:	:	:
07-kim-ai	214	88	(	(	-LRB-
07-kim-ai	214	89	i	i	NN
07-kim-ai	214	90	)	)	-RRB-
07-kim-ai	214	91	intellectual	intellectual	JJ
07-kim-ai	214	92	isolation	isolation	NN
07-kim-ai	214	93	and	and	CC
07-kim-ai	214	94	bigotry	bigotry	NN
07-kim-ai	214	95	hampering	hamper	VBG
07-kim-ai	214	96	civic	civic	JJ
07-kim-ai	214	97	discourse	discourse	NN
07-kim-ai	214	98	;	;	:
07-kim-ai	214	99	(	(	-LRB-
07-kim-ai	214	100	ii	ii	NN
07-kim-ai	214	101	)	)	-RRB-
07-kim-ai	214	102	weakening	weaken	VBG
07-kim-ai	214	103	of	of	IN
07-kim-ai	214	104	cognitive	cognitive	JJ
07-kim-ai	214	105	agency	agency	NN
07-kim-ai	214	106	and	and	CC
07-kim-ai	214	107	autonomy	autonomy	NN
07-kim-ai	214	108	;	;	,
07-kim-ai	214	109	(	(	-LRB-
07-kim-ai	214	110	iii	iii	NNP
07-kim-ai	214	111	)	)	-RRB-
07-kim-ai	214	112	social	social	JJ
07-kim-ai	214	113	impact	impact	NN
07-kim-ai	214	114	of	of	IN
07-kim-ai	214	115	scholarship	scholarship	NN
07-kim-ai	214	116	and	and	CC
07-kim-ai	214	117	research	research	NN
07-kim-ai	214	118	based	base	VBN
07-kim-ai	214	119	upon	upon	IN
07-kim-ai	214	120	flawed	flawed	JJ
07-kim-ai	214	121	AI	AI	NNP
07-kim-ai	214	122	algorithms	algorithm	NNS
07-kim-ai	214	123	.	.	.
07-kim-ai	215	1	In	in	IN
07-kim-ai	215	2	the	the	DT
07-kim-ai	215	3	near	near	JJ
07-kim-ai	215	4	future	future	NN
07-kim-ai	215	5	,	,	,
07-kim-ai	215	6	libraries	library	NNS
07-kim-ai	215	7	will	will	MD
07-kim-ai	215	8	be	be	VB
07-kim-ai	215	9	acquiring	acquire	VBG
07-kim-ai	215	10	,	,	,
07-kim-ai	215	11	building	build	VBG
07-kim-ai	215	12	,	,	,
07-kim-ai	215	13	customizing	customizing	NN
07-kim-ai	215	14	,	,	,
07-kim-ai	215	15	and	and	CC
07-kim-ai	215	16	implementing	implement	VBG
07-kim-ai	215	17	many	many	JJ
07-kim-ai	215	18	personalized	personalize	VBN
07-kim-ai	215	19	and	and	CC
07-kim-ai	215	20	automated	automate	VBD
07-kim-ai	215	21	information	information	NN
07-kim-ai	215	22	systems	system	NNS
07-kim-ai	215	23	.	.	.
07-kim-ai	216	1	Given	give	VBN
07-kim-ai	216	2	this	this	DT
07-kim-ai	216	3	,	,	,
07-kim-ai	216	4	the	the	DT
07-kim-ai	216	5	challenges	challenge	NNS
07-kim-ai	216	6	related	relate	VBN
07-kim-ai	216	7	to	to	IN
07-kim-ai	216	8	the	the	DT
07-kim-ai	216	9	AI-	AI-	NNP
07-kim-ai	216	10	powered	power	VBN
07-kim-ai	216	11	automated	automated	JJ
07-kim-ai	216	12	information	information	NN
07-kim-ai	216	13	environment	environment	NN
07-kim-ai	216	14	are	be	VBP
07-kim-ai	216	15	highly	highly	RB
07-kim-ai	216	16	relevant	relevant	JJ
07-kim-ai	216	17	to	to	IN
07-kim-ai	216	18	them	-PRON-	PRP
07-kim-ai	216	19	.	.	.
07-kim-ai	217	1	At	at	IN
07-kim-ai	217	2	present	present	NN
07-kim-ai	217	3	,	,	,
07-kim-ai	217	4	libraries	library	NNS
07-kim-ai	217	5	are	be	VBP
07-kim-ai	217	6	at	at	IN
07-kim-ai	217	7	an	an	DT
07-kim-ai	217	8	early	early	JJ
07-kim-ai	217	9	stage	stage	NN
07-kim-ai	217	10	in	in	IN
07-kim-ai	217	11	developing	develop	VBG
07-kim-ai	217	12	AI	ai	NN
07-kim-ai	217	13	applications	application	NNS
07-kim-ai	217	14	and	and	CC
07-kim-ai	217	15	applying	apply	VBG
07-kim-ai	217	16	machine	machine	NN
07-kim-ai	217	17	learning	learning	NN
07-kim-ai	217	18	and	and	CC
07-kim-ai	217	19	deep	deep	JJ
07-kim-ai	217	20	learn-	learn-	JJ
07-kim-ai	217	21	ing	ing	NN
07-kim-ai	217	22	techniques	technique	NNS
07-kim-ai	217	23	to	to	TO
07-kim-ai	217	24	improve	improve	VB
07-kim-ai	217	25	library	library	JJ
07-kim-ai	217	26	services	service	NNS
07-kim-ai	217	27	,	,	,
07-kim-ai	217	28	systems	system	NNS
07-kim-ai	217	29	,	,	,
07-kim-ai	217	30	and	and	CC
07-kim-ai	217	31	operations	operation	NNS
07-kim-ai	217	32	.	.	.
07-kim-ai	218	1	But	but	CC
07-kim-ai	218	2	the	the	DT
07-kim-ai	218	3	general	general	JJ
07-kim-ai	218	4	issues	issue	NNS
07-kim-ai	218	5	of	of	IN
07-kim-ai	218	6	hidden	hide	VBN
07-kim-ai	218	7	biases	bias	NNS
07-kim-ai	218	8	and	and	CC
07-kim-ai	218	9	the	the	DT
07-kim-ai	218	10	lack	lack	NN
07-kim-ai	218	11	of	of	IN
07-kim-ai	218	12	explainability	explainability	NN
07-kim-ai	218	13	in	in	IN
07-kim-ai	218	14	machine	machine	NN
07-kim-ai	218	15	learning	learning	NN
07-kim-ai	218	16	and	and	CC
07-kim-ai	218	17	deep	deep	JJ
07-kim-ai	218	18	learning	learning	NN
07-kim-ai	218	19	are	be	VBP
07-kim-ai	218	20	already	already	RB
07-kim-ai	218	21	gaining	gain	VBG
07-kim-ai	218	22	awareness	awareness	NN
07-kim-ai	218	23	in	in	IN
07-kim-ai	218	24	the	the	DT
07-kim-ai	218	25	library	library	NN
07-kim-ai	218	26	community	community	NN
07-kim-ai	218	27	.	.	.
07-kim-ai	219	1	As	as	IN
07-kim-ai	219	2	we	-PRON-	PRP
07-kim-ai	219	3	have	have	VBP
07-kim-ai	219	4	seen	see	VBN
07-kim-ai	219	5	in	in	IN
07-kim-ai	219	6	the	the	DT
07-kim-ai	219	7	trolley	trolley	NN
07-kim-ai	219	8	problem	problem	NN
07-kim-ai	219	9	,	,	,
07-kim-ai	219	10	whether	whether	IN
07-kim-ai	219	11	a	a	DT
07-kim-ai	219	12	certain	certain	JJ
07-kim-ai	219	13	action	action	NN
07-kim-ai	219	14	is	be	VBZ
07-kim-ai	219	15	moral	moral	JJ
07-kim-ai	219	16	is	be	VBZ
07-kim-ai	219	17	not	not	RB
07-kim-ai	219	18	a	a	DT
07-kim-ai	219	19	line	line	NN
07-kim-ai	219	20	that	that	WDT
07-kim-ai	219	21	can	can	MD
07-kim-ai	219	22	be	be	VB
07-kim-ai	219	23	drawn	draw	VBN
07-kim-ai	219	24	with	with	IN
07-kim-ai	219	25	absolute	absolute	JJ
07-kim-ai	219	26	clarity	clarity	NN
07-kim-ai	219	27	.	.	.
07-kim-ai	220	1	It	-PRON-	PRP
07-kim-ai	220	2	is	be	VBZ
07-kim-ai	220	3	entirely	entirely	RB
07-kim-ai	220	4	possible	possible	JJ
07-kim-ai	220	5	for	for	IN
07-kim-ai	220	6	fully	fully	RB
07-kim-ai	220	7	-	-	HYPH
07-kim-ai	220	8	functioning	function	VBG
07-kim-ai	220	9	moral	moral	JJ
07-kim-ai	220	10	agents	agent	NNS
07-kim-ai	220	11	to	to	TO
07-kim-ai	220	12	make	make	VB
07-kim-ai	220	13	different	different	JJ
07-kim-ai	220	14	judgements	judgement	NNS
07-kim-ai	220	15	.	.	.
07-kim-ai	221	1	In	in	IN
07-kim-ai	221	2	addition	addition	NN
07-kim-ai	221	3	,	,	,
07-kim-ai	221	4	there	there	EX
07-kim-ai	221	5	is	be	VBZ
07-kim-ai	221	6	the	the	DT
07-kim-ai	221	7	matter	matter	NN
07-kim-ai	221	8	of	of	IN
07-kim-ai	221	9	morality	morality	NN
07-kim-ai	221	10	that	that	IN
07-kim-ai	221	11	our	-PRON-	PRP$
07-kim-ai	221	12	tools	tool	NNS
07-kim-ai	221	13	and	and	CC
07-kim-ai	221	14	systems	system	NNS
07-kim-ai	221	15	display	display	NN
07-kim-ai	221	16	.	.	.
07-kim-ai	222	1	This	this	DT
07-kim-ai	222	2	is	be	VBZ
07-kim-ai	222	3	called	call	VBN
07-kim-ai	222	4	“	"	``
07-kim-ai	222	5	machine	machine	NN
07-kim-ai	222	6	morality	morality	NN
07-kim-ai	222	7	”	"	''
07-kim-ai	222	8	in	in	IN
07-kim-ai	222	9	relation	relation	NN
07-kim-ai	222	10	to	to	IN
07-kim-ai	222	11	AI	AI	NNP
07-kim-ai	222	12	systems	system	NNS
07-kim-ai	222	13	.	.	.
07-kim-ai	223	1	Wallach	Wallach	NNP
07-kim-ai	223	2	and	and	CC
07-kim-ai	223	3	Allen	Allen	NNP
07-kim-ai	223	4	(	(	-LRB-
07-kim-ai	223	5	2009	2009	CD
07-kim-ai	223	6	)	)	-RRB-
07-kim-ai	223	7	argue	argue	VBP
07-kim-ai	223	8	that	that	IN
07-kim-ai	223	9	there	there	EX
07-kim-ai	223	10	are	be	VBP
07-kim-ai	223	11	three	three	CD
07-kim-ai	223	12	distinct	distinct	JJ
07-kim-ai	223	13	levels	level	NNS
07-kim-ai	223	14	of	of	IN
07-kim-ai	223	15	machine	machine	NN
07-kim-ai	223	16	morality	morality	NN
07-kim-ai	223	17	:	:	:
07-kim-ai	223	18	oper-	oper-	NNP
07-kim-ai	223	19	ational	ational	JJ
07-kim-ai	223	20	morality	morality	NN
07-kim-ai	223	21	,	,	,
07-kim-ai	223	22	functional	functional	JJ
07-kim-ai	223	23	morality	morality	NN
07-kim-ai	223	24	,	,	,
07-kim-ai	223	25	and	and	CC
07-kim-ai	223	26	full	full	JJ
07-kim-ai	223	27	moral	moral	JJ
07-kim-ai	223	28	agency	agency	NN
07-kim-ai	223	29	(	(	-LRB-
07-kim-ai	223	30	26	26	CD
07-kim-ai	223	31	)	)	-RRB-
07-kim-ai	223	32	.	.	.
07-kim-ai	224	1	Operational	operational	JJ
07-kim-ai	224	2	morality	morality	NN
07-kim-ai	224	3	is	be	VBZ
07-kim-ai	224	4	found	find	VBN
07-kim-ai	224	5	in	in	IN
07-kim-ai	224	6	systems	system	NNS
07-kim-ai	224	7	that	that	WDT
07-kim-ai	224	8	are	be	VBP
07-kim-ai	224	9	low	low	JJ
07-kim-ai	224	10	in	in	IN
07-kim-ai	224	11	both	both	CC
07-kim-ai	224	12	autonomy	autonomy	NN
07-kim-ai	224	13	and	and	CC
07-kim-ai	224	14	ethical	ethical	JJ
07-kim-ai	224	15	sensitivity	sensitivity	NN
07-kim-ai	224	16	.	.	.
07-kim-ai	225	1	At	at	IN
07-kim-ai	225	2	this	this	DT
07-kim-ai	225	3	level	level	NN
07-kim-ai	225	4	of	of	IN
07-kim-ai	225	5	machine	machine	NN
07-kim-ai	225	6	moral-	moral-	NN
07-kim-ai	225	7	ity	ity	NN
07-kim-ai	225	8	,	,	,
07-kim-ai	225	9	a	a	DT
07-kim-ai	225	10	machine	machine	NN
07-kim-ai	225	11	or	or	CC
07-kim-ai	225	12	a	a	DT
07-kim-ai	225	13	tool	tool	NN
07-kim-ai	225	14	is	be	VBZ
07-kim-ai	225	15	given	give	VBN
07-kim-ai	225	16	a	a	DT
07-kim-ai	225	17	mechanism	mechanism	NN
07-kim-ai	225	18	that	that	WDT
07-kim-ai	225	19	prevents	prevent	VBZ
07-kim-ai	225	20	its	-PRON-	PRP$
07-kim-ai	225	21	immoral	immoral	JJ
07-kim-ai	225	22	use	use	NN
07-kim-ai	225	23	,	,	,
07-kim-ai	225	24	but	but	CC
07-kim-ai	225	25	the	the	DT
07-kim-ai	225	26	mechanism	mechanism	NN
07-kim-ai	225	27	24See	24see	CD
07-kim-ai	225	28	Tashea	Tashea	NNP
07-kim-ai	225	29	2017	2017	CD
07-kim-ai	225	30	and	and	CC
07-kim-ai	225	31	Stanley	Stanley	NNP
07-kim-ai	225	32	2017	2017	CD
07-kim-ai	225	33	.	.	.
07-kim-ai	226	1	25This	25this	CD
07-kim-ai	226	2	is	be	VBZ
07-kim-ai	226	3	by	by	IN
07-kim-ai	226	4	no	no	DT
07-kim-ai	226	5	means	means	NN
07-kim-ai	226	6	an	an	DT
07-kim-ai	226	7	exhaustive	exhaustive	JJ
07-kim-ai	226	8	list	list	NN
07-kim-ai	226	9	.	.	.
07-kim-ai	227	1	User	user	NN
07-kim-ai	227	2	privacy	privacy	NN
07-kim-ai	227	3	and	and	CC
07-kim-ai	227	4	potential	potential	JJ
07-kim-ai	227	5	surveillance	surveillance	NN
07-kim-ai	227	6	are	be	VBP
07-kim-ai	227	7	examples	example	NNS
07-kim-ai	227	8	of	of	IN
07-kim-ai	227	9	other	other	JJ
07-kim-ai	227	10	important	important	JJ
07-kim-ai	227	11	ethical	ethical	JJ
07-kim-ai	227	12	challenges	challenge	NNS
07-kim-ai	227	13	,	,	,
07-kim-ai	227	14	which	which	WDT
07-kim-ai	227	15	I	-PRON-	PRP
07-kim-ai	227	16	do	do	VBP
07-kim-ai	227	17	not	not	RB
07-kim-ai	227	18	discuss	discuss	VB
07-kim-ai	227	19	here	here	RB
07-kim-ai	227	20	.	.	.
07-kim-ai	228	1	82	82	CD
07-kim-ai	228	2	Machine	Machine	NNP
07-kim-ai	228	3	Learning	Learning	NNP
07-kim-ai	228	4	,	,	,
07-kim-ai	228	5	Libraries	Libraries	NNPS
07-kim-ai	228	6	,	,	,
07-kim-ai	228	7	and	and	CC
07-kim-ai	228	8	Cross	Cross	NNP
07-kim-ai	228	9	-	-	NNP
07-kim-ai	228	10	Disciplinary	Disciplinary	NNP
07-kim-ai	228	11	ResearchǔChapter	ResearchǔChapter	NNP
07-kim-ai	228	12	7	7	CD
07-kim-ai	228	13	is	be	VBZ
07-kim-ai	228	14	within	within	IN
07-kim-ai	228	15	the	the	DT
07-kim-ai	228	16	full	full	JJ
07-kim-ai	228	17	control	control	NN
07-kim-ai	228	18	of	of	IN
07-kim-ai	228	19	the	the	DT
07-kim-ai	228	20	user	user	NN
07-kim-ai	228	21	.	.	.
07-kim-ai	229	1	Such	such	JJ
07-kim-ai	229	2	operational	operational	JJ
07-kim-ai	229	3	morality	morality	NN
07-kim-ai	229	4	exists	exist	VBZ
07-kim-ai	229	5	in	in	IN
07-kim-ai	229	6	a	a	DT
07-kim-ai	229	7	gun	gun	NN
07-kim-ai	229	8	with	with	IN
07-kim-ai	229	9	a	a	DT
07-kim-ai	229	10	childproof	childproof	JJ
07-kim-ai	229	11	safety	safety	NN
07-kim-ai	229	12	mechanism	mechanism	NN
07-kim-ai	229	13	,	,	,
07-kim-ai	229	14	for	for	IN
07-kim-ai	229	15	example	example	NN
07-kim-ai	229	16	.	.	.
07-kim-ai	230	1	A	a	DT
07-kim-ai	230	2	gun	gun	NN
07-kim-ai	230	3	with	with	IN
07-kim-ai	230	4	a	a	DT
07-kim-ai	230	5	safety	safety	NN
07-kim-ai	230	6	mechanism	mechanism	NN
07-kim-ai	230	7	is	be	VBZ
07-kim-ai	230	8	neither	neither	CC
07-kim-ai	230	9	autonomous	autonomous	JJ
07-kim-ai	230	10	nor	nor	CC
07-kim-ai	230	11	sen-	sen-	RB
07-kim-ai	230	12	sitive	sitive	JJ
07-kim-ai	230	13	to	to	IN
07-kim-ai	230	14	ethical	ethical	JJ
07-kim-ai	230	15	concerns	concern	NNS
07-kim-ai	230	16	related	relate	VBN
07-kim-ai	230	17	to	to	IN
07-kim-ai	230	18	its	-PRON-	PRP$
07-kim-ai	230	19	use	use	NN
07-kim-ai	230	20	.	.	.
07-kim-ai	231	1	By	by	IN
07-kim-ai	231	2	contrast	contrast	NN
07-kim-ai	231	3	,	,	,
07-kim-ai	231	4	machines	machine	NNS
07-kim-ai	231	5	with	with	IN
07-kim-ai	231	6	functional	functional	JJ
07-kim-ai	231	7	morality	morality	NN
07-kim-ai	231	8	do	do	VBP
07-kim-ai	231	9	possess	possess	VB
07-kim-ai	231	10	a	a	DT
07-kim-ai	231	11	certain	certain	JJ
07-kim-ai	231	12	level	level	NN
07-kim-ai	231	13	of	of	IN
07-kim-ai	231	14	autonomy	autonomy	NN
07-kim-ai	231	15	and	and	CC
07-kim-ai	231	16	ethical	ethical	JJ
07-kim-ai	231	17	sensitivity	sensitivity	NN
07-kim-ai	231	18	.	.	.
07-kim-ai	232	1	This	this	DT
07-kim-ai	232	2	category	category	NN
07-kim-ai	232	3	includes	include	VBZ
07-kim-ai	232	4	AI	AI	NNP
07-kim-ai	232	5	systems	system	NNS
07-kim-ai	232	6	with	with	IN
07-kim-ai	232	7	significant	significant	JJ
07-kim-ai	232	8	autonomy	autonomy	NN
07-kim-ai	232	9	and	and	CC
07-kim-ai	232	10	little	little	JJ
07-kim-ai	232	11	ethical	ethical	JJ
07-kim-ai	232	12	sensitivity	sensitivity	NN
07-kim-ai	232	13	or	or	CC
07-kim-ai	232	14	those	those	DT
07-kim-ai	232	15	with	with	IN
07-kim-ai	232	16	little	little	JJ
07-kim-ai	232	17	autonomy	autonomy	NN
07-kim-ai	232	18	and	and	CC
07-kim-ai	232	19	high	high	JJ
07-kim-ai	232	20	ethical	ethical	JJ
07-kim-ai	232	21	sensitivity	sensitivity	NN
07-kim-ai	232	22	.	.	.
07-kim-ai	233	1	An	an	DT
07-kim-ai	233	2	autonomous	autonomous	JJ
07-kim-ai	233	3	drone	drone	NN
07-kim-ai	233	4	would	would	MD
07-kim-ai	233	5	fall	fall	VB
07-kim-ai	233	6	under	under	IN
07-kim-ai	233	7	the	the	DT
07-kim-ai	233	8	former	former	JJ
07-kim-ai	233	9	type	type	NN
07-kim-ai	233	10	,	,	,
07-kim-ai	233	11	while	while	IN
07-kim-ai	233	12	MedEthEx	medethex	NN
07-kim-ai	233	13	,	,	,
07-kim-ai	233	14	an	an	DT
07-kim-ai	233	15	ethical	ethical	JJ
07-kim-ai	233	16	decision	decision	NN
07-kim-ai	233	17	-	-	HYPH
07-kim-ai	233	18	support	support	NN
07-kim-ai	233	19	AI	AI	NNP
07-kim-ai	233	20	recommendation	recommendation	NN
07-kim-ai	233	21	system	system	NN
07-kim-ai	233	22	for	for	IN
07-kim-ai	233	23	clinicians	clinician	NNS
07-kim-ai	233	24	,	,	,
07-kim-ai	233	25	would	would	MD
07-kim-ai	233	26	be	be	VB
07-kim-ai	233	27	of	of	IN
07-kim-ai	233	28	the	the	DT
07-kim-ai	233	29	latter	latter	JJ
07-kim-ai	233	30	.	.	.
07-kim-ai	234	1	Lastly	lastly	RB
07-kim-ai	234	2	,	,	,
07-kim-ai	234	3	Wallach	Wallach	NNP
07-kim-ai	234	4	and	and	CC
07-kim-ai	234	5	Allen	Allen	NNP
07-kim-ai	234	6	regard	regard	NN
07-kim-ai	234	7	systems	system	NNS
07-kim-ai	234	8	with	with	IN
07-kim-ai	234	9	high	high	JJ
07-kim-ai	234	10	autonomy	autonomy	NN
07-kim-ai	234	11	and	and	CC
07-kim-ai	234	12	high	high	JJ
07-kim-ai	234	13	ethical	ethical	JJ
07-kim-ai	234	14	sensitivity	sensitivity	NN
07-kim-ai	234	15	as	as	IN
07-kim-ai	234	16	having	have	VBG
07-kim-ai	234	17	full	full	JJ
07-kim-ai	234	18	moral	moral	JJ
07-kim-ai	234	19	agency	agency	NN
07-kim-ai	234	20	,	,	,
07-kim-ai	234	21	as	as	RB
07-kim-ai	234	22	much	much	RB
07-kim-ai	234	23	as	as	IN
07-kim-ai	234	24	humans	human	NNS
07-kim-ai	234	25	do	do	VBP
07-kim-ai	234	26	.	.	.
07-kim-ai	235	1	This	this	DT
07-kim-ai	235	2	means	mean	VBZ
07-kim-ai	235	3	that	that	IN
07-kim-ai	235	4	those	those	DT
07-kim-ai	235	5	systems	system	NNS
07-kim-ai	235	6	would	would	MD
07-kim-ai	235	7	have	have	VB
07-kim-ai	235	8	a	a	DT
07-kim-ai	235	9	mental	mental	JJ
07-kim-ai	235	10	rep-	rep-	NN
07-kim-ai	235	11	resentation	resentation	NN
07-kim-ai	235	12	of	of	IN
07-kim-ai	235	13	values	value	NNS
07-kim-ai	235	14	and	and	CC
07-kim-ai	235	15	the	the	DT
07-kim-ai	235	16	capacity	capacity	NN
07-kim-ai	235	17	for	for	IN
07-kim-ai	235	18	moral	moral	JJ
07-kim-ai	235	19	reasoning	reasoning	NN
07-kim-ai	235	20	.	.	.
07-kim-ai	236	1	Such	such	JJ
07-kim-ai	236	2	machines	machine	NNS
07-kim-ai	236	3	can	can	MD
07-kim-ai	236	4	be	be	VB
07-kim-ai	236	5	held	hold	VBN
07-kim-ai	236	6	morally	morally	RB
07-kim-ai	236	7	responsible	responsible	JJ
07-kim-ai	236	8	for	for	IN
07-kim-ai	236	9	their	-PRON-	PRP$
07-kim-ai	236	10	actions	action	NNS
07-kim-ai	236	11	.	.	.
07-kim-ai	237	1	We	-PRON-	PRP
07-kim-ai	237	2	do	do	VBP
07-kim-ai	237	3	not	not	RB
07-kim-ai	237	4	know	know	VB
07-kim-ai	237	5	whether	whether	IN
07-kim-ai	237	6	AI	AI	NNP
07-kim-ai	237	7	will	will	MD
07-kim-ai	237	8	be	be	VB
07-kim-ai	237	9	able	able	JJ
07-kim-ai	237	10	to	to	TO
07-kim-ai	237	11	produce	produce	VB
07-kim-ai	237	12	such	such	PDT
07-kim-ai	237	13	a	a	DT
07-kim-ai	237	14	machine	machine	NN
07-kim-ai	237	15	with	with	IN
07-kim-ai	237	16	full	full	JJ
07-kim-ai	237	17	moral	moral	JJ
07-kim-ai	237	18	agency	agency	NN
07-kim-ai	237	19	.	.	.
07-kim-ai	238	1	If	if	IN
07-kim-ai	238	2	the	the	DT
07-kim-ai	238	3	current	current	JJ
07-kim-ai	238	4	direction	direction	NN
07-kim-ai	238	5	to	to	TO
07-kim-ai	238	6	automate	automate	VB
07-kim-ai	238	7	more	more	JJR
07-kim-ai	238	8	and	and	CC
07-kim-ai	238	9	more	more	JJR
07-kim-ai	238	10	human	human	JJ
07-kim-ai	238	11	tasks	task	NNS
07-kim-ai	238	12	for	for	IN
07-kim-ai	238	13	cost	cost	NN
07-kim-ai	238	14	savings	saving	NNS
07-kim-ai	238	15	and	and	CC
07-kim-ai	238	16	efficiency	efficiency	NN
07-kim-ai	238	17	at	at	IN
07-kim-ai	238	18	scale	scale	NN
07-kim-ai	238	19	continues	continue	VBZ
07-kim-ai	238	20	,	,	,
07-kim-ai	238	21	however	however	RB
07-kim-ai	238	22	,	,	,
07-kim-ai	238	23	most	most	JJS
07-kim-ai	238	24	of	of	IN
07-kim-ai	238	25	the	the	DT
07-kim-ai	238	26	more	more	RBR
07-kim-ai	238	27	sophisticated	sophisticated	JJ
07-kim-ai	238	28	AI	AI	NNP
07-kim-ai	238	29	applications	application	NNS
07-kim-ai	238	30	to	to	TO
07-kim-ai	238	31	come	come	VB
07-kim-ai	238	32	will	will	MD
07-kim-ai	238	33	be	be	VB
07-kim-ai	238	34	of	of	IN
07-kim-ai	238	35	the	the	DT
07-kim-ai	238	36	kind	kind	NN
07-kim-ai	238	37	with	with	IN
07-kim-ai	238	38	functional	functional	JJ
07-kim-ai	238	39	morality	morality	NN
07-kim-ai	238	40	,	,	,
07-kim-ai	238	41	particularly	particularly	RB
07-kim-ai	238	42	the	the	DT
07-kim-ai	238	43	kind	kind	NN
07-kim-ai	238	44	that	that	WDT
07-kim-ai	238	45	combines	combine	VBZ
07-kim-ai	238	46	a	a	DT
07-kim-ai	238	47	relatively	relatively	RB
07-kim-ai	238	48	high	high	JJ
07-kim-ai	238	49	level	level	NN
07-kim-ai	238	50	of	of	IN
07-kim-ai	238	51	autonomy	autonomy	NN
07-kim-ai	238	52	and	and	CC
07-kim-ai	238	53	a	a	DT
07-kim-ai	238	54	lower	low	JJR
07-kim-ai	238	55	level	level	NN
07-kim-ai	238	56	of	of	IN
07-kim-ai	238	57	ethical	ethical	JJ
07-kim-ai	238	58	sensitivity	sensitivity	NN
07-kim-ai	238	59	.	.	.
07-kim-ai	239	1	In	in	IN
07-kim-ai	239	2	the	the	DT
07-kim-ai	239	3	beginning	beginning	NN
07-kim-ai	239	4	of	of	IN
07-kim-ai	239	5	this	this	DT
07-kim-ai	239	6	chapter	chapter	NN
07-kim-ai	239	7	,	,	,
07-kim-ai	239	8	I	-PRON-	PRP
07-kim-ai	239	9	mentioned	mention	VBD
07-kim-ai	239	10	that	that	IN
07-kim-ai	239	11	the	the	DT
07-kim-ai	239	12	goal	goal	NN
07-kim-ai	239	13	of	of	IN
07-kim-ai	239	14	AI	AI	NNP
07-kim-ai	239	15	is	be	VBZ
07-kim-ai	239	16	to	to	TO
07-kim-ai	239	17	create	create	VB
07-kim-ai	239	18	an	an	DT
07-kim-ai	239	19	artificial	artificial	JJ
07-kim-ai	239	20	system	system	NN
07-kim-ai	239	21	—	—	:
07-kim-ai	239	22	whether	whether	IN
07-kim-ai	239	23	it	-PRON-	PRP
07-kim-ai	239	24	be	be	VB
07-kim-ai	239	25	a	a	DT
07-kim-ai	239	26	piece	piece	NN
07-kim-ai	239	27	of	of	IN
07-kim-ai	239	28	software	software	NN
07-kim-ai	239	29	or	or	CC
07-kim-ai	239	30	a	a	DT
07-kim-ai	239	31	machine	machine	NN
07-kim-ai	239	32	with	with	IN
07-kim-ai	239	33	a	a	DT
07-kim-ai	239	34	physical	physical	JJ
07-kim-ai	239	35	body	body	NN
07-kim-ai	239	36	—	—	:
07-kim-ai	239	37	that	that	DT
07-kim-ai	239	38	is	be	VBZ
07-kim-ai	239	39	as	as	RB
07-kim-ai	239	40	intelligent	intelligent	JJ
07-kim-ai	239	41	as	as	IN
07-kim-ai	239	42	a	a	DT
07-kim-ai	239	43	human	human	NN
07-kim-ai	239	44	in	in	IN
07-kim-ai	239	45	its	-PRON-	PRP$
07-kim-ai	239	46	performance	performance	NN
07-kim-ai	239	47	,	,	,
07-kim-ai	239	48	either	either	CC
07-kim-ai	239	49	broadly	broadly	RB
07-kim-ai	239	50	in	in	IN
07-kim-ai	239	51	all	all	DT
07-kim-ai	239	52	areas	area	NNS
07-kim-ai	239	53	of	of	IN
07-kim-ai	239	54	human	human	JJ
07-kim-ai	239	55	activities	activity	NNS
07-kim-ai	239	56	or	or	CC
07-kim-ai	239	57	narrowly	narrowly	RB
07-kim-ai	239	58	in	in	IN
07-kim-ai	239	59	a	a	DT
07-kim-ai	239	60	specific	specific	JJ
07-kim-ai	239	61	activity	activity	NN
07-kim-ai	239	62	.	.	.
07-kim-ai	240	1	But	but	CC
07-kim-ai	240	2	what	what	WP
07-kim-ai	240	3	does	do	VBZ
07-kim-ai	240	4	“	"	``
07-kim-ai	240	5	as	as	RB
07-kim-ai	240	6	intelligent	intelligent	JJ
07-kim-ai	240	7	as	as	IN
07-kim-ai	240	8	a	a	DT
07-kim-ai	240	9	human	human	NN
07-kim-ai	240	10	”	"	''
07-kim-ai	240	11	exactly	exactly	RB
07-kim-ai	240	12	mean	mean	VB
07-kim-ai	240	13	?	?	.
07-kim-ai	241	1	If	if	IN
07-kim-ai	241	2	morality	morality	NN
07-kim-ai	241	3	is	be	VBZ
07-kim-ai	241	4	an	an	DT
07-kim-ai	241	5	integral	integral	JJ
07-kim-ai	241	6	component	component	NN
07-kim-ai	241	7	of	of	IN
07-kim-ai	241	8	human	human	JJ
07-kim-ai	241	9	-	-	HYPH
07-kim-ai	241	10	level	level	NN
07-kim-ai	241	11	intelligence	intelligence	NN
07-kim-ai	241	12	,	,	,
07-kim-ai	241	13	AI	AI	NNP
07-kim-ai	241	14	research	research	NN
07-kim-ai	241	15	needs	need	VBZ
07-kim-ai	241	16	to	to	TO
07-kim-ai	241	17	pay	pay	VB
07-kim-ai	241	18	more	more	JJR
07-kim-ai	241	19	attention	attention	NN
07-kim-ai	241	20	to	to	IN
07-kim-ai	241	21	intelligence	intelligence	NN
07-kim-ai	241	22	not	not	RB
07-kim-ai	241	23	only	only	RB
07-kim-ai	241	24	in	in	IN
07-kim-ai	241	25	accomplishing	accomplish	VBG
07-kim-ai	241	26	a	a	DT
07-kim-ai	241	27	goal	goal	NN
07-kim-ai	241	28	but	but	CC
07-kim-ai	241	29	also	also	RB
07-kim-ai	241	30	in	in	IN
07-kim-ai	241	31	doing	do	VBG
07-kim-ai	241	32	so	so	RB
07-kim-ai	241	33	ethically.26	ethically.26	NNP
07-kim-ai	241	34	In	in	IN
07-kim-ai	241	35	that	that	DT
07-kim-ai	241	36	light	light	NN
07-kim-ai	241	37	,	,	,
07-kim-ai	241	38	it	-PRON-	PRP
07-kim-ai	241	39	is	be	VBZ
07-kim-ai	241	40	meaningful	meaningful	JJ
07-kim-ai	241	41	to	to	TO
07-kim-ai	241	42	ask	ask	VB
07-kim-ai	241	43	what	what	WDT
07-kim-ai	241	44	level	level	NN
07-kim-ai	241	45	of	of	IN
07-kim-ai	241	46	autonomy	autonomy	NN
07-kim-ai	241	47	and	and	CC
07-kim-ai	241	48	ethical	ethical	JJ
07-kim-ai	241	49	sensitivity	sensitivity	NN
07-kim-ai	241	50	a	a	DT
07-kim-ai	241	51	given	give	VBN
07-kim-ai	241	52	AI	ai	NN
07-kim-ai	241	53	system	system	NN
07-kim-ai	241	54	is	be	VBZ
07-kim-ai	241	55	equipped	equip	VBN
07-kim-ai	241	56	with	with	IN
07-kim-ai	241	57	,	,	,
07-kim-ai	241	58	and	and	CC
07-kim-ai	241	59	what	what	WDT
07-kim-ai	241	60	level	level	NN
07-kim-ai	241	61	of	of	IN
07-kim-ai	241	62	machine	machine	NN
07-kim-ai	241	63	morality	morality	NN
07-kim-ai	241	64	is	be	VBZ
07-kim-ai	241	65	appropriate	appropriate	JJ
07-kim-ai	241	66	for	for	IN
07-kim-ai	241	67	its	-PRON-	PRP$
07-kim-ai	241	68	purpose	purpose	NN
07-kim-ai	241	69	.	.	.
07-kim-ai	242	1	In	in	IN
07-kim-ai	242	2	designing	design	VBG
07-kim-ai	242	3	an	an	DT
07-kim-ai	242	4	AI	AI	NNP
07-kim-ai	242	5	system	system	NN
07-kim-ai	242	6	,	,	,
07-kim-ai	242	7	it	-PRON-	PRP
07-kim-ai	242	8	would	would	MD
07-kim-ai	242	9	be	be	VB
07-kim-ai	242	10	helpful	helpful	JJ
07-kim-ai	242	11	to	to	TO
07-kim-ai	242	12	consider	consider	VB
07-kim-ai	242	13	what	what	WDT
07-kim-ai	242	14	level	level	NN
07-kim-ai	242	15	of	of	IN
07-kim-ai	242	16	autonomy	autonomy	NN
07-kim-ai	242	17	and	and	CC
07-kim-ai	242	18	ethical	ethical	JJ
07-kim-ai	242	19	sensitivity	sensitivity	NN
07-kim-ai	242	20	would	would	MD
07-kim-ai	242	21	be	be	VB
07-kim-ai	242	22	best	well	RBS
07-kim-ai	242	23	suited	suit	VBN
07-kim-ai	242	24	for	for	IN
07-kim-ai	242	25	its	-PRON-	PRP$
07-kim-ai	242	26	purpose	purpose	NN
07-kim-ai	242	27	and	and	CC
07-kim-ai	242	28	whether	whether	IN
07-kim-ai	242	29	it	-PRON-	PRP
07-kim-ai	242	30	is	be	VBZ
07-kim-ai	242	31	feasible	feasible	JJ
07-kim-ai	242	32	to	to	TO
07-kim-ai	242	33	provide	provide	VB
07-kim-ai	242	34	that	that	DT
07-kim-ai	242	35	level	level	NN
07-kim-ai	242	36	of	of	IN
07-kim-ai	242	37	machine	machine	NN
07-kim-ai	242	38	morality	morality	NN
07-kim-ai	242	39	for	for	IN
07-kim-ai	242	40	the	the	DT
07-kim-ai	242	41	system	system	NN
07-kim-ai	242	42	in	in	IN
07-kim-ai	242	43	question	question	NN
07-kim-ai	242	44	.	.	.
07-kim-ai	243	1	In	in	IN
07-kim-ai	243	2	general	general	JJ
07-kim-ai	243	3	,	,	,
07-kim-ai	243	4	the	the	DT
07-kim-ai	243	5	narrower	narrow	JJR
07-kim-ai	243	6	the	the	DT
07-kim-ai	243	7	function	function	NN
07-kim-ai	243	8	or	or	CC
07-kim-ai	243	9	the	the	DT
07-kim-ai	243	10	do-	do-	JJ
07-kim-ai	243	11	main	main	NN
07-kim-ai	243	12	of	of	IN
07-kim-ai	243	13	an	an	DT
07-kim-ai	243	14	AI	AI	NNP
07-kim-ai	243	15	system	system	NN
07-kim-ai	243	16	is	be	VBZ
07-kim-ai	243	17	,	,	,
07-kim-ai	243	18	the	the	DT
07-kim-ai	243	19	easier	easy	JJR
07-kim-ai	243	20	it	-PRON-	PRP
07-kim-ai	243	21	will	will	MD
07-kim-ai	243	22	be	be	VB
07-kim-ai	243	23	to	to	TO
07-kim-ai	243	24	equip	equip	VB
07-kim-ai	243	25	it	-PRON-	PRP
07-kim-ai	243	26	with	with	IN
07-kim-ai	243	27	an	an	DT
07-kim-ai	243	28	appropriate	appropriate	JJ
07-kim-ai	243	29	level	level	NN
07-kim-ai	243	30	of	of	IN
07-kim-ai	243	31	autonomy	autonomy	NN
07-kim-ai	243	32	and	and	CC
07-kim-ai	243	33	ethical	ethical	JJ
07-kim-ai	243	34	sensitivity	sensitivity	NN
07-kim-ai	243	35	.	.	.
07-kim-ai	244	1	In	in	IN
07-kim-ai	244	2	evaluating	evaluate	VBG
07-kim-ai	244	3	and	and	CC
07-kim-ai	244	4	designing	design	VBG
07-kim-ai	244	5	an	an	DT
07-kim-ai	244	6	AI	AI	NNP
07-kim-ai	244	7	system	system	NN
07-kim-ai	244	8	,	,	,
07-kim-ai	244	9	it	-PRON-	PRP
07-kim-ai	244	10	will	will	MD
07-kim-ai	244	11	be	be	VB
07-kim-ai	244	12	important	important	JJ
07-kim-ai	244	13	to	to	TO
07-kim-ai	244	14	test	test	VB
07-kim-ai	244	15	the	the	DT
07-kim-ai	244	16	actual	actual	JJ
07-kim-ai	244	17	outcome	outcome	NN
07-kim-ai	244	18	against	against	IN
07-kim-ai	244	19	the	the	DT
07-kim-ai	244	20	anticipated	anticipated	JJ
07-kim-ai	244	21	outcome	outcome	NN
07-kim-ai	244	22	in	in	IN
07-kim-ai	244	23	different	different	JJ
07-kim-ai	244	24	types	type	NNS
07-kim-ai	244	25	of	of	IN
07-kim-ai	244	26	cases	case	NNS
07-kim-ai	244	27	in	in	IN
07-kim-ai	244	28	order	order	NN
07-kim-ai	244	29	to	to	TO
07-kim-ai	244	30	identify	identify	VB
07-kim-ai	244	31	potential	potential	JJ
07-kim-ai	244	32	problems	problem	NNS
07-kim-ai	244	33	.	.	.
07-kim-ai	245	1	System	system	NN
07-kim-ai	245	2	-	-	HYPH
07-kim-ai	245	3	wide	wide	JJ
07-kim-ai	245	4	audits	audits	NN
07-kim-ai	245	5	to	to	TO
07-kim-ai	245	6	detect	detect	VB
07-kim-ai	245	7	well	well	RB
07-kim-ai	245	8	-	-	HYPH
07-kim-ai	245	9	known	know	VBN
07-kim-ai	245	10	biases	bias	NNS
07-kim-ai	245	11	,	,	,
07-kim-ai	245	12	such	such	JJ
07-kim-ai	245	13	as	as	IN
07-kim-ai	245	14	gender	gender	NN
07-kim-ai	245	15	discrimina-	discrimina-	NNP
07-kim-ai	245	16	tion	tion	NN
07-kim-ai	245	17	or	or	CC
07-kim-ai	245	18	racism	racism	NN
07-kim-ai	245	19	,	,	,
07-kim-ai	245	20	can	can	MD
07-kim-ai	245	21	serve	serve	VB
07-kim-ai	245	22	as	as	IN
07-kim-ai	245	23	an	an	DT
07-kim-ai	245	24	effective	effective	JJ
07-kim-ai	245	25	strategy.27	strategy.27	NNP
07-kim-ai	245	26	Other	other	JJ
07-kim-ai	245	27	undetected	undetected	JJ
07-kim-ai	245	28	problems	problem	NNS
07-kim-ai	245	29	may	may	MD
07-kim-ai	245	30	surface	surface	VB
07-kim-ai	245	31	only	only	RB
07-kim-ai	245	32	after	after	IN
07-kim-ai	245	33	the	the	DT
07-kim-ai	245	34	AI	AI	NNP
07-kim-ai	245	35	system	system	NN
07-kim-ai	245	36	is	be	VBZ
07-kim-ai	245	37	deployed	deploy	VBN
07-kim-ai	245	38	.	.	.
07-kim-ai	246	1	Having	have	VBG
07-kim-ai	246	2	a	a	DT
07-kim-ai	246	3	mechanism	mechanism	NN
07-kim-ai	246	4	to	to	TO
07-kim-ai	246	5	continually	continually	RB
07-kim-ai	246	6	test	test	VB
07-kim-ai	246	7	an	an	DT
07-kim-ai	246	8	AI	AI	NNP
07-kim-ai	246	9	algorithm	algorithm	NN
07-kim-ai	246	10	to	to	TO
07-kim-ai	246	11	iden-	iden-	VB
07-kim-ai	246	12	tify	tify	VB
07-kim-ai	246	13	those	those	DT
07-kim-ai	246	14	unnoticed	unnoticed	JJ
07-kim-ai	246	15	problems	problem	NNS
07-kim-ai	246	16	and	and	CC
07-kim-ai	246	17	feeding	feed	VBG
07-kim-ai	246	18	the	the	DT
07-kim-ai	246	19	test	test	NN
07-kim-ai	246	20	result	result	VB
07-kim-ai	246	21	back	back	RB
07-kim-ai	246	22	into	into	IN
07-kim-ai	246	23	the	the	DT
07-kim-ai	246	24	algorithm	algorithm	NN
07-kim-ai	246	25	for	for	IN
07-kim-ai	246	26	retraining	retraining	NN
07-kim-ai	246	27	will	will	MD
07-kim-ai	246	28	be	be	VB
07-kim-ai	246	29	another	another	DT
07-kim-ai	246	30	way	way	NN
07-kim-ai	246	31	to	to	TO
07-kim-ai	246	32	deal	deal	VB
07-kim-ai	246	33	with	with	IN
07-kim-ai	246	34	algorithmic	algorithmic	JJ
07-kim-ai	246	35	biases	bias	NNS
07-kim-ai	246	36	.	.	.
07-kim-ai	247	1	Those	those	DT
07-kim-ai	247	2	who	who	WP
07-kim-ai	247	3	build	build	VBP
07-kim-ai	247	4	AI	AI	NNP
07-kim-ai	247	5	systems	system	NNS
07-kim-ai	247	6	will	will	MD
07-kim-ai	247	7	also	also	RB
07-kim-ai	247	8	benefit	benefit	VB
07-kim-ai	247	9	from	from	IN
07-kim-ai	247	10	consulting	consult	VBG
07-kim-ai	247	11	existing	exist	VBG
07-kim-ai	247	12	principles	principle	NNS
07-kim-ai	247	13	and	and	CC
07-kim-ai	247	14	guidelines	guideline	NNS
07-kim-ai	247	15	such	such	JJ
07-kim-ai	247	16	as	as	IN
07-kim-ai	247	17	FAT	FAT	NNP
07-kim-ai	247	18	/	/	SYM
07-kim-ai	247	19	ML	ML	NNP
07-kim-ai	247	20	’s	’s	POS
07-kim-ai	247	21	“	"	``
07-kim-ai	247	22	Principles	principle	NNS
07-kim-ai	247	23	for	for	IN
07-kim-ai	247	24	Accountable	Accountable	NNP
07-kim-ai	247	25	Algorithms	Algorithms	NNP
07-kim-ai	247	26	and	and	CC
07-kim-ai	247	27	a	a	DT
07-kim-ai	247	28	Social	Social	NNP
07-kim-ai	247	29	Impact	Impact	NNP
07-kim-ai	247	30	Statement	Statement	NNP
07-kim-ai	247	31	for	for	IN
07-kim-ai	247	32	Algorithms	Algorithms	NNP
07-kim-ai	247	33	.	.	.
07-kim-ai	247	34	”28	”28	''
07-kim-ai	247	35	We	-PRON-	PRP
07-kim-ai	247	36	may	may	MD
07-kim-ai	247	37	also	also	RB
07-kim-ai	247	38	want	want	VB
07-kim-ai	247	39	to	to	TO
07-kim-ai	247	40	rethink	rethink	VB
07-kim-ai	247	41	how	how	WRB
07-kim-ai	247	42	and	and	CC
07-kim-ai	247	43	where	where	WRB
07-kim-ai	247	44	we	-PRON-	PRP
07-kim-ai	247	45	apply	apply	VBP
07-kim-ai	247	46	AI	AI	NNP
07-kim-ai	247	47	.	.	.
07-kim-ai	248	1	We	-PRON-	PRP
07-kim-ai	248	2	and	and	CC
07-kim-ai	248	3	our	-PRON-	PRP$
07-kim-ai	248	4	society	society	NN
07-kim-ai	248	5	do	do	VBP
07-kim-ai	248	6	not	not	RB
07-kim-ai	248	7	have	have	VB
07-kim-ai	248	8	26Here	26here	CD
07-kim-ai	248	9	,	,	,
07-kim-ai	248	10	I	-PRON-	PRP
07-kim-ai	248	11	regard	regard	VBP
07-kim-ai	248	12	intelligence	intelligence	NN
07-kim-ai	248	13	as	as	IN
07-kim-ai	248	14	the	the	DT
07-kim-ai	248	15	ability	ability	NN
07-kim-ai	248	16	to	to	TO
07-kim-ai	248	17	accomplish	accomplish	VB
07-kim-ai	248	18	complex	complex	JJ
07-kim-ai	248	19	goals	goal	NNS
07-kim-ai	248	20	following	follow	VBG
07-kim-ai	248	21	Tegmark	Tegmark	NNP
07-kim-ai	248	22	2017	2017	CD
07-kim-ai	248	23	.	.	.
07-kim-ai	249	1	For	for	IN
07-kim-ai	249	2	more	more	JJR
07-kim-ai	249	3	discussion	discussion	NN
07-kim-ai	249	4	on	on	IN
07-kim-ai	249	5	intelligence	intelligence	NN
07-kim-ai	249	6	and	and	CC
07-kim-ai	249	7	goals	goal	NNS
07-kim-ai	249	8	,	,	,
07-kim-ai	249	9	see	see	VB
07-kim-ai	249	10	Chapter	chapter	NN
07-kim-ai	249	11	2	2	CD
07-kim-ai	249	12	and	and	CC
07-kim-ai	249	13	Chapter	chapter	NN
07-kim-ai	249	14	7	7	CD
07-kim-ai	249	15	.	.	.
07-kim-ai	250	1	27These	27these	JJ
07-kim-ai	250	2	audits	audits	NN
07-kim-ai	250	3	are	be	VBP
07-kim-ai	250	4	far	far	RB
07-kim-ai	250	5	from	from	IN
07-kim-ai	250	6	foolproof	foolproof	JJ
07-kim-ai	250	7	,	,	,
07-kim-ai	250	8	but	but	CC
07-kim-ai	250	9	the	the	DT
07-kim-ai	250	10	detection	detection	NN
07-kim-ai	250	11	of	of	IN
07-kim-ai	250	12	hidden	hidden	JJ
07-kim-ai	250	13	biases	bias	NNS
07-kim-ai	250	14	will	will	MD
07-kim-ai	250	15	be	be	VB
07-kim-ai	250	16	crucial	crucial	JJ
07-kim-ai	250	17	in	in	IN
07-kim-ai	250	18	making	make	VBG
07-kim-ai	250	19	AI	AI	NNP
07-kim-ai	250	20	algorithms	algorithm	NNS
07-kim-ai	250	21	more	more	RBR
07-kim-ai	250	22	accountable	accountable	JJ
07-kim-ai	250	23	and	and	CC
07-kim-ai	250	24	their	-PRON-	PRP$
07-kim-ai	250	25	decisions	decision	NNS
07-kim-ai	250	26	more	more	RBR
07-kim-ai	250	27	ethical	ethical	JJ
07-kim-ai	250	28	.	.	.
07-kim-ai	251	1	A	a	DT
07-kim-ai	251	2	debiasing	debiasing	JJ
07-kim-ai	251	3	algorithm	algorithm	NN
07-kim-ai	251	4	can	can	MD
07-kim-ai	251	5	also	also	RB
07-kim-ai	251	6	be	be	VB
07-kim-ai	251	7	used	use	VBN
07-kim-ai	251	8	during	during	IN
07-kim-ai	251	9	the	the	DT
07-kim-ai	251	10	training	training	NN
07-kim-ai	251	11	stage	stage	NN
07-kim-ai	251	12	of	of	IN
07-kim-ai	251	13	an	an	DT
07-kim-ai	251	14	AI	AI	NNP
07-kim-ai	251	15	algorithm	algorithm	NN
07-kim-ai	251	16	to	to	TO
07-kim-ai	251	17	reduce	reduce	VB
07-kim-ai	251	18	hidden	hide	VBN
07-kim-ai	251	19	biases	bias	NNS
07-kim-ai	251	20	in	in	IN
07-kim-ai	251	21	training	training	NN
07-kim-ai	251	22	data	datum	NNS
07-kim-ai	251	23	.	.	.
07-kim-ai	252	1	See	see	VB
07-kim-ai	252	2	Amini	Amini	NNP
07-kim-ai	252	3	et	et	NNP
07-kim-ai	252	4	al	al	NNP
07-kim-ai	252	5	.	.	.
07-kim-ai	253	1	2019	2019	CD
07-kim-ai	253	2	,	,	,
07-kim-ai	253	3	Knight	Knight	NNP
07-kim-ai	253	4	2019b	2019b	CD
07-kim-ai	253	5	,	,	,
07-kim-ai	253	6	and	and	CC
07-kim-ai	253	7	Courtland	Courtland	NNP
07-kim-ai	253	8	2018	2018	CD
07-kim-ai	253	9	.	.	.
07-kim-ai	254	1	28See	28see	CD
07-kim-ai	254	2	?	?	.
07-kim-ai	254	3	iiTb	iitb	UH
07-kim-ai	254	4	,	,	,
07-kim-ai	254	5	ffrrrX7	ffrrrx7	CD
07-kim-ai	254	6	�	�	NNP
07-kim-ai	254	7	iKHXQ`;f`2bQm`+2bfT`BM+BTH2b@7Q`@	iKHXQ`;f`2bQm`+2bfT`BM+BTH2b@7Q`@	NNP
07-kim-ai	254	8	�	�	NNP
07-kim-ai	254	9	++QmMi	++qmmi	XX
07-kim-ai	254	10	�	�	JJ
07-kim-ai	254	11	#H2@	#H2@	NNP
07-kim-ai	254	12	�	�	NNP
07-kim-ai	254	13	H;Q`Bi?Kb	h;q`bi?kb	ADD
07-kim-ai	254	14	.	.	.
07-kim-ai	255	1	Other	other	JJ
07-kim-ai	255	2	principles	principle	NNS
07-kim-ai	255	3	and	and	CC
07-kim-ai	255	4	guidelines	guideline	NNS
07-kim-ai	255	5	include	include	VBP
07-kim-ai	255	6	“	"	``
07-kim-ai	255	7	Ethics	Ethics	NNP
07-kim-ai	255	8	Guidelines	Guidelines	NNPS
07-kim-ai	255	9	for	for	IN
07-kim-ai	255	10	Trustworthy	Trustworthy	NNP
07-kim-ai	255	11	AI	AI	NNP
07-kim-ai	255	12	”	"	''
07-kim-ai	255	13	(	(	-LRB-
07-kim-ai	255	14	?	?	.
07-kim-ai	255	15	iiTb	iitb	UH
07-kim-ai	255	16	,	,	,
07-kim-ai	255	17	ff2+X2m`QT	ff2+X2m`QT	NNP
07-kim-ai	255	18	�	�	NNP
07-kim-ai	255	19	X2mf	X2mf	NNP
07-kim-ai	255	20	/	/	SYM
07-kim-ai	255	21	B;Bi	B;Bi	NNP
07-kim-ai	255	22	�	�	NNP
07-kim-ai	255	23	H@b	H@b	NNP
07-kim-ai	255	24	BM;H2@K	BM;H2@K	NNP
07-kim-ai	255	25	�	�	NNP
07-kim-ai	255	26	`F2if2MfM2rbf2i?B+b@;mB/2HBM2b@i`mbirQ`i?v@	`f2if2mfm2rbf2i?b+b@;mb/2hbm2b@i`mbirq`i?v@	NN
07-kim-ai	255	27	�	�	NNP
07-kim-ai	255	28	B	b	NN
07-kim-ai	255	29	)	)	-RRB-
07-kim-ai	255	30	and	and	CC
07-kim-ai	255	31	“	"	``
07-kim-ai	255	32	Algorithmic	Algorithmic	NNP
07-kim-ai	255	33	Impact	Impact	NNP
07-kim-ai	255	34	Assessments	assessment	NNS
07-kim-ai	255	35	:	:	:
07-kim-ai	255	36	A	a	DT
07-kim-ai	255	37	Practical	Practical	NNP
07-kim-ai	255	38	Framework	Framework	NNP
07-kim-ai	255	39	For	for	IN
07-kim-ai	255	40	Public	Public	NNP
07-kim-ai	255	41	Agency	Agency	NNP
07-kim-ai	255	42	Accountability	Accountability	NNP
07-kim-ai	255	43	”	"	''
07-kim-ai	255	44	(	(	-LRB-
07-kim-ai	255	45	?	?	.
07-kim-ai	255	46	iiTb	iitb	UH
07-kim-ai	255	47	,	,	,
07-kim-ai	255	48	ff	ff	VB
07-kim-ai	255	49	�	�	NNP
07-kim-ai	255	50	BMQrBMbiBimi2XQ`;f	BMQrBMbiBimi2XQ`;f	NNP
07-kim-ai	255	51	�	�	NNP
07-kim-ai	255	52	B	B	NNP
07-kim-ai	255	53	�	�	NNP
07-kim-ai	255	54	`2TQ`ikyR3XT	`2TQ`ikyR3XT	NNP
07-kim-ai	255	55	/7	/7	.
07-kim-ai	255	56	)	)	-RRB-
07-kim-ai	255	57	.	.	.
07-kim-ai	256	1	https://www.fatml.org/resources/principles-for-accountable-algorithms	https://www.fatml.org/resources/principles-for-accountable-algorithms	NNP
07-kim-ai	256	2	https://ec.europa.eu/digital-single-market/en/news/ethics-guidelines-trustworthy-ai	https://ec.europa.eu/digital-single-market/en/news/ethics-guidelines-trustworthy-ai	NNP
07-kim-ai	256	3	https://ec.europa.eu/digital-single-market/en/news/ethics-guidelines-trustworthy-ai	https://ec.europa.eu/digital-single-market/en/news/ethics-guidelines-trustworthy-ai	NNP
07-kim-ai	256	4	https://ainowinstitute.org/aiareport2018.pdf	https://ainowinstitute.org/aiareport2018.pdf	NNP
07-kim-ai	256	5	https://ainowinstitute.org/aiareport2018.pdf	https://ainowinstitute.org/aiareport2018.pdf	NNP
07-kim-ai	256	6	Kim	Kim	NNP
07-kim-ai	256	7	83	83	CD
07-kim-ai	256	8	to	to	TO
07-kim-ai	256	9	use	use	VB
07-kim-ai	256	10	AI	AI	NNP
07-kim-ai	256	11	to	to	TO
07-kim-ai	256	12	equip	equip	VB
07-kim-ai	256	13	all	all	PDT
07-kim-ai	256	14	our	-PRON-	PRP$
07-kim-ai	256	15	systems	system	NNS
07-kim-ai	256	16	and	and	CC
07-kim-ai	256	17	machines	machine	NNS
07-kim-ai	256	18	with	with	IN
07-kim-ai	256	19	human-	human-	NN
07-kim-ai	256	20	or	or	CC
07-kim-ai	256	21	superhuman	superhuman	JJ
07-kim-ai	256	22	-	-	HYPH
07-kim-ai	256	23	level	level	NN
07-kim-ai	256	24	performance	performance	NN
07-kim-ai	256	25	.	.	.
07-kim-ai	257	1	This	this	DT
07-kim-ai	257	2	is	be	VBZ
07-kim-ai	257	3	particularly	particularly	RB
07-kim-ai	257	4	so	so	RB
07-kim-ai	257	5	if	if	IN
07-kim-ai	257	6	the	the	DT
07-kim-ai	257	7	pursuit	pursuit	NN
07-kim-ai	257	8	of	of	IN
07-kim-ai	257	9	such	such	JJ
07-kim-ai	257	10	human-	human-	NN
07-kim-ai	257	11	or	or	CC
07-kim-ai	257	12	superhuman	superhuman	JJ
07-kim-ai	257	13	-	-	HYPH
07-kim-ai	257	14	level	level	NN
07-kim-ai	257	15	performance	performance	NN
07-kim-ai	257	16	is	be	VBZ
07-kim-ai	257	17	likely	likely	JJ
07-kim-ai	257	18	to	to	TO
07-kim-ai	257	19	increase	increase	VB
07-kim-ai	257	20	unethical	unethical	JJ
07-kim-ai	257	21	decisions	decision	NNS
07-kim-ai	257	22	that	that	WDT
07-kim-ai	257	23	negatively	negatively	RB
07-kim-ai	257	24	impact	impact	VBP
07-kim-ai	257	25	a	a	DT
07-kim-ai	257	26	significant	significant	JJ
07-kim-ai	257	27	number	number	NN
07-kim-ai	257	28	of	of	IN
07-kim-ai	257	29	people	people	NNS
07-kim-ai	257	30	.	.	.
07-kim-ai	258	1	We	-PRON-	PRP
07-kim-ai	258	2	do	do	VBP
07-kim-ai	258	3	not	not	RB
07-kim-ai	258	4	have	have	VB
07-kim-ai	258	5	to	to	TO
07-kim-ai	258	6	task	task	NN
07-kim-ai	258	7	AI	AI	NNP
07-kim-ai	258	8	with	with	IN
07-kim-ai	258	9	always	always	RB
07-kim-ai	258	10	automating	automate	VBG
07-kim-ai	258	11	away	away	RP
07-kim-ai	258	12	human	human	JJ
07-kim-ai	258	13	work	work	NN
07-kim-ai	258	14	and	and	CC
07-kim-ai	258	15	decisions	decision	NNS
07-kim-ai	258	16	as	as	RB
07-kim-ai	258	17	much	much	RB
07-kim-ai	258	18	as	as	IN
07-kim-ai	258	19	possible	possible	JJ
07-kim-ai	258	20	.	.	.
07-kim-ai	259	1	What	what	WP
07-kim-ai	259	2	if	if	IN
07-kim-ai	259	3	we	-PRON-	PRP
07-kim-ai	259	4	reframe	reframe	VBP
07-kim-ai	259	5	AI	AI	NNP
07-kim-ai	259	6	’s	’s	POS
07-kim-ai	259	7	role	role	NN
07-kim-ai	259	8	as	as	IN
07-kim-ai	259	9	helping	help	VBG
07-kim-ai	259	10	people	people	NNS
07-kim-ai	259	11	become	become	VB
07-kim-ai	259	12	more	more	RBR
07-kim-ai	259	13	intelligent	intelligent	JJ
07-kim-ai	259	14	and	and	CC
07-kim-ai	259	15	more	more	RBR
07-kim-ai	259	16	capable	capable	JJ
07-kim-ai	259	17	where	where	WRB
07-kim-ai	259	18	they	-PRON-	PRP
07-kim-ai	259	19	struggle	struggle	VBP
07-kim-ai	259	20	or	or	CC
07-kim-ai	259	21	experience	experience	NN
07-kim-ai	259	22	disadvantages	disadvantage	NNS
07-kim-ai	259	23	,	,	,
07-kim-ai	259	24	such	such	JJ
07-kim-ai	259	25	as	as	IN
07-kim-ai	259	26	critical	critical	JJ
07-kim-ai	259	27	thinking	thinking	NN
07-kim-ai	259	28	,	,	,
07-kim-ai	259	29	civic	civic	JJ
07-kim-ai	259	30	participation	participation	NN
07-kim-ai	259	31	,	,	,
07-kim-ai	259	32	healthy	healthy	JJ
07-kim-ai	259	33	liv-	liv-	VBD
07-kim-ai	259	34	ing	ing	JJ
07-kim-ai	259	35	,	,	,
07-kim-ai	259	36	financial	financial	JJ
07-kim-ai	259	37	literacy	literacy	NN
07-kim-ai	259	38	,	,	,
07-kim-ai	259	39	dyslexia	dyslexia	NNP
07-kim-ai	259	40	,	,	,
07-kim-ai	259	41	or	or	CC
07-kim-ai	259	42	hearing	hearing	NN
07-kim-ai	259	43	loss	loss	NN
07-kim-ai	259	44	?	?	.
07-kim-ai	260	1	What	what	WDT
07-kim-ai	260	2	kind	kind	NN
07-kim-ai	260	3	of	of	IN
07-kim-ai	260	4	AI	ai	NN
07-kim-ai	260	5	-	-	HYPH
07-kim-ai	260	6	driven	drive	VBN
07-kim-ai	260	7	information	information	NN
07-kim-ai	260	8	systems	system	NNS
07-kim-ai	260	9	and	and	CC
07-kim-ai	260	10	environments	environment	NNS
07-kim-ai	260	11	would	would	MD
07-kim-ai	260	12	be	be	VB
07-kim-ai	260	13	created	create	VBN
07-kim-ai	260	14	if	if	IN
07-kim-ai	260	15	libraries	library	NNS
07-kim-ai	260	16	approach	approach	VBP
07-kim-ai	260	17	AI	AI	NNP
07-kim-ai	260	18	with	with	IN
07-kim-ai	260	19	such	such	JJ
07-kim-ai	260	20	intention	intention	NN
07-kim-ai	260	21	from	from	IN
07-kim-ai	260	22	the	the	DT
07-kim-ai	260	23	beginning	beginning	NN
07-kim-ai	260	24	?	?	.
07-kim-ai	261	1	References	References	NNP
07-kim-ai	261	2	Alexander	Alexander	NNP
07-kim-ai	261	3	,	,	,
07-kim-ai	261	4	Larry	Larry	NNP
07-kim-ai	261	5	,	,	,
07-kim-ai	261	6	and	and	CC
07-kim-ai	261	7	Michael	Michael	NNP
07-kim-ai	261	8	Moore	Moore	NNP
07-kim-ai	261	9	.	.	.
07-kim-ai	262	1	2016	2016	CD
07-kim-ai	262	2	.	.	.
07-kim-ai	263	1	“	"	``
07-kim-ai	263	2	Deontological	Deontological	NNP
07-kim-ai	263	3	Ethics	Ethics	NNPS
07-kim-ai	263	4	.	.	.
07-kim-ai	263	5	”	"	''
07-kim-ai	263	6	In	in	IN
07-kim-ai	263	7	The	the	DT
07-kim-ai	263	8	Stanford	Stanford	NNP
07-kim-ai	263	9	Encyclo-	Encyclo-	NNP
07-kim-ai	263	10	pedia	pedia	NN
07-kim-ai	263	11	of	of	IN
07-kim-ai	263	12	Philosophy	Philosophy	NNP
07-kim-ai	263	13	,	,	,
07-kim-ai	263	14	edited	edit	VBN
07-kim-ai	263	15	by	by	IN
07-kim-ai	263	16	Edward	Edward	NNP
07-kim-ai	263	17	N.	N.	NNP
07-kim-ai	263	18	Zalta	Zalta	NNP
07-kim-ai	263	19	,	,	,
07-kim-ai	263	20	Winter	Winter	NNP
07-kim-ai	263	21	2016	2016	CD
07-kim-ai	263	22	.	.	.
07-kim-ai	264	1	Metaphysics	Metaphysics	NNP
07-kim-ai	264	2	Research	Research	NNP
07-kim-ai	264	3	Lab	Lab	NNP
07-kim-ai	264	4	,	,	,
07-kim-ai	264	5	Stanford	Stanford	NNP
07-kim-ai	264	6	University	University	NNP
07-kim-ai	264	7	.	.	.
07-kim-ai	265	1	?	?	.
07-kim-ai	265	2	iiTb	iitb	UH
07-kim-ai	265	3	,	,	,
07-kim-ai	265	4	ffTH	ffTH	NNP
07-kim-ai	265	5	�	�	NNP
07-kim-ai	265	6	iQXbi	iQXbi	NNS
07-kim-ai	265	7	�	�	NNP
07-kim-ai	265	8	M7Q`/X2	m7q`/x2	NN
07-kim-ai	265	9	/	/	SYM
07-kim-ai	265	10	mf	mf	IN
07-kim-ai	265	11	�	�	NNP
07-kim-ai	265	12	`+?Bp2bfrBMkyRef2Mi`B2	`+?Bp2bfrBMkyRef2Mi`B2	NNP
07-kim-ai	265	13	bf2i?B+b@/2QMiQHQ;B+	bf2i?b+b@/2qmiqhq;b+	NN
07-kim-ai	265	14	�	�	NNP
07-kim-ai	265	15	Hf	Hf	NNP
07-kim-ai	265	16	.	.	.
07-kim-ai	266	1	Amini	Amini	NNP
07-kim-ai	266	2	,	,	,
07-kim-ai	266	3	Alexander	Alexander	NNP
07-kim-ai	266	4	,	,	,
07-kim-ai	266	5	Ava	Ava	NNP
07-kim-ai	266	6	P.	P.	NNP
07-kim-ai	266	7	Soleimany	Soleimany	NNP
07-kim-ai	266	8	,	,	,
07-kim-ai	266	9	Wilko	Wilko	NNP
07-kim-ai	266	10	Schwarting	Schwarting	NNP
07-kim-ai	266	11	,	,	,
07-kim-ai	266	12	Sangeeta	Sangeeta	NNP
07-kim-ai	266	13	N.	N.	NNP
07-kim-ai	266	14	Bhatia	Bhatia	NNP
07-kim-ai	266	15	,	,	,
07-kim-ai	266	16	and	and	CC
07-kim-ai	266	17	Daniela	Daniela	NNP
07-kim-ai	266	18	Rus	Rus	NNP
07-kim-ai	266	19	.	.	.
07-kim-ai	267	1	2019	2019	CD
07-kim-ai	267	2	.	.	.
07-kim-ai	268	1	“	"	``
07-kim-ai	268	2	Uncovering	uncover	VBG
07-kim-ai	268	3	and	and	CC
07-kim-ai	268	4	Mitigating	Mitigating	NNP
07-kim-ai	268	5	Algorithmic	Algorithmic	NNP
07-kim-ai	268	6	Bias	Bias	NNP
07-kim-ai	268	7	through	through	IN
07-kim-ai	268	8	Learned	learn	VBN
07-kim-ai	268	9	Latent	Latent	NNP
07-kim-ai	268	10	Structure	Structure	NNP
07-kim-ai	268	11	.	.	.
07-kim-ai	268	12	”	"	''
07-kim-ai	268	13	In	in	IN
07-kim-ai	268	14	Proceedings	Proceedings	NNP
07-kim-ai	268	15	of	of	IN
07-kim-ai	268	16	the	the	DT
07-kim-ai	268	17	2019	2019	CD
07-kim-ai	268	18	AAAI	AAAI	NNP
07-kim-ai	268	19	/	/	SYM
07-kim-ai	268	20	ACM	ACM	NNP
07-kim-ai	268	21	Conference	Conference	NNP
07-kim-ai	268	22	on	on	IN
07-kim-ai	268	23	AI	AI	NNP
07-kim-ai	268	24	,	,	,
07-kim-ai	268	25	Ethics	Ethics	NNP
07-kim-ai	268	26	,	,	,
07-kim-ai	268	27	and	and	CC
07-kim-ai	268	28	Society	Society	NNP
07-kim-ai	268	29	,	,	,
07-kim-ai	268	30	289–295	289–295	CD
07-kim-ai	268	31	.	.	.
07-kim-ai	269	1	AIES	AIES	NNP
07-kim-ai	269	2	’	’	POS
07-kim-ai	269	3	19	19	CD
07-kim-ai	269	4	.	.	.
07-kim-ai	270	1	New	New	NNP
07-kim-ai	270	2	York	York	NNP
07-kim-ai	270	3	,	,	,
07-kim-ai	270	4	NY	NY	NNP
07-kim-ai	270	5	,	,	,
07-kim-ai	270	6	USA	USA	NNP
07-kim-ai	270	7	:	:	:
07-kim-ai	270	8	Association	Association	NNP
07-kim-ai	270	9	for	for	IN
07-kim-ai	270	10	Computing	Computing	NNP
07-kim-ai	270	11	Machinery	Machinery	NNP
07-kim-ai	270	12	.	.	.
07-kim-ai	271	1	?	?	.
07-kim-ai	271	2	iiTb	iitb	UH
07-kim-ai	271	3	,	,	,
07-kim-ai	271	4	ff	ff	NNP
07-kim-ai	271	5	/	/	SYM
07-kim-ai	271	6	QBXQ`;f	QBXQ`;f	NNP
07-kim-ai	271	7	RyXRR98fjjyeeR3XjjR9k9j	RyXRR98fjjyeeR3XjjR9k9j	NNS
07-kim-ai	271	8	.	.	.
07-kim-ai	272	1	Averkamp	Averkamp	NNP
07-kim-ai	272	2	,	,	,
07-kim-ai	272	3	Shawn	Shawn	NNP
07-kim-ai	272	4	,	,	,
07-kim-ai	272	5	and	and	CC
07-kim-ai	272	6	Julie	Julie	NNP
07-kim-ai	272	7	Hardesty	Hardesty	NNP
07-kim-ai	272	8	.	.	.
07-kim-ai	273	1	2020	2020	CD
07-kim-ai	273	2	.	.	.
07-kim-ai	274	1	“	"	``
07-kim-ai	274	2	AI	AI	NNP
07-kim-ai	274	3	Is	be	VBZ
07-kim-ai	274	4	Such	such	PDT
07-kim-ai	274	5	a	a	DT
07-kim-ai	274	6	Tool	tool	NN
07-kim-ai	274	7	:	:	:
07-kim-ai	274	8	Keeping	keep	VBG
07-kim-ai	274	9	Your	-PRON-	PRP$
07-kim-ai	274	10	Machine	machine	NN
07-kim-ai	274	11	Learn-	Learn-	NNP
07-kim-ai	274	12	ing	ing	NNP
07-kim-ai	274	13	Outputs	Outputs	NNP
07-kim-ai	274	14	in	in	IN
07-kim-ai	274	15	Check	Check	NNP
07-kim-ai	274	16	.	.	.
07-kim-ai	274	17	”	"	''
07-kim-ai	274	18	Presented	present	VBN
07-kim-ai	274	19	at	at	IN
07-kim-ai	274	20	the	the	DT
07-kim-ai	274	21	Code4lib	Code4lib	NNP
07-kim-ai	274	22	Conference	Conference	NNP
07-kim-ai	274	23	,	,	,
07-kim-ai	274	24	Pittsburgh	Pittsburgh	NNP
07-kim-ai	274	25	,	,	,
07-kim-ai	274	26	PA	PA	NNP
07-kim-ai	274	27	,	,	,
07-kim-ai	274	28	March	March	NNP
07-kim-ai	274	29	11	11	CD
07-kim-ai	274	30	.	.	.
07-kim-ai	275	1	?	?	.
07-kim-ai	275	2	iiTb	iitb	UH
07-kim-ai	275	3	,	,	,
07-kim-ai	275	4	ffkykyX+Q/29HB#XQ`;fi	ffkykyx+q/29hb#xq`;fi	JJ
07-kim-ai	275	5	�	�	NN
07-kim-ai	275	6	HFbf	hfbf	UH
07-kim-ai	275	7	�	�	NNP
07-kim-ai	275	8	A@Bb@bm+	a@bb@bm+	NN
07-kim-ai	275	9	?	?	.
07-kim-ai	276	1	@	@	NFP
07-kim-ai	276	2	�	�	NNP
07-kim-ai	276	3	@iQQH@E22TBM;@vQm`@K	@iQQH@E22TBM;@vQm`@K	NNP
07-kim-ai	276	4	�	�	NNP
07-kim-ai	276	5	+	+	CC
07-kim-ai	276	6	?	?	.
07-kim-ai	276	7	BM2@H2	bm2@h2	JJ
07-kim-ai	276	8	�	�	NNP
07-kim-ai	276	9	`MBM;@QmiTmib@BM@+?2+F.	`mbm;@qmitmib@bm@+?2+f.	CD
07-kim-ai	277	1	Blewer	Blewer	NNP
07-kim-ai	277	2	,	,	,
07-kim-ai	277	3	Ashley	Ashley	NNP
07-kim-ai	277	4	,	,	,
07-kim-ai	277	5	Bohyun	Bohyun	NNP
07-kim-ai	277	6	Kim	Kim	NNP
07-kim-ai	277	7	,	,	,
07-kim-ai	277	8	and	and	CC
07-kim-ai	277	9	Eric	Eric	NNP
07-kim-ai	277	10	Phetteplace	Phetteplace	NNP
07-kim-ai	277	11	.	.	.
07-kim-ai	278	1	2018	2018	CD
07-kim-ai	278	2	.	.	.
07-kim-ai	279	1	“	"	``
07-kim-ai	279	2	Reflections	reflection	NNS
07-kim-ai	279	3	on	on	IN
07-kim-ai	279	4	Code4Lib	Code4Lib	NNP
07-kim-ai	279	5	2018	2018	CD
07-kim-ai	279	6	.	.	.
07-kim-ai	279	7	”	"	''
07-kim-ai	279	8	ACRL	ACRL	NNP
07-kim-ai	279	9	TechConnect	TechConnect	NNP
07-kim-ai	279	10	(	(	-LRB-
07-kim-ai	279	11	blog	blog	NN
07-kim-ai	279	12	)	)	-RRB-
07-kim-ai	279	13	.	.	.
07-kim-ai	280	1	March	March	NNP
07-kim-ai	280	2	12	12	CD
07-kim-ai	280	3	,	,	,
07-kim-ai	280	4	2018	2018	CD
07-kim-ai	280	5	.	.	.
07-kim-ai	281	1	?	?	.
07-kim-ai	281	2	iiTb	iitb	UH
07-kim-ai	281	3	,	,	,
07-kim-ai	281	4	ff	ff	VB
07-kim-ai	281	5	�	�	NNP
07-kim-ai	281	6	+`HX	+`HX	.
07-kim-ai	281	7	�	�	JJ
07-kim-ai	281	8	H	H	NNP
07-kim-ai	281	9	�	�	NNP
07-kim-ai	281	10	XQ`;fi2+?+QMM2+i	XQ`;fi2+?+QMM2+i	NNP
07-kim-ai	281	11	fTQbif`27H2+iBQMb@QM@+Q/29HB#@kyR3f	ftqbif`27h2+ibqmb@qm@+q/29hb#@kyr3f	CD
07-kim-ai	281	12	.	.	.
07-kim-ai	282	1	Boden	Boden	NNP
07-kim-ai	282	2	,	,	,
07-kim-ai	282	3	Margaret	Margaret	NNP
07-kim-ai	282	4	A.	A.	NNP
07-kim-ai	283	1	2016	2016	CD
07-kim-ai	283	2	.	.	.
07-kim-ai	284	1	AI	ai	NN
07-kim-ai	284	2	:	:	:
07-kim-ai	284	3	Its	-PRON-	PRP$
07-kim-ai	284	4	Nature	Nature	NNP
07-kim-ai	284	5	and	and	CC
07-kim-ai	284	6	Future	Future	NNP
07-kim-ai	284	7	.	.	.
07-kim-ai	285	1	Oxford	oxford	NN
07-kim-ai	285	2	:	:	:
07-kim-ai	285	3	Oxford	Oxford	NNP
07-kim-ai	285	4	University	University	NNP
07-kim-ai	285	5	Press	Press	NNP
07-kim-ai	285	6	.	.	.
07-kim-ai	286	1	Bonnefon	Bonnefon	NNP
07-kim-ai	286	2	,	,	,
07-kim-ai	286	3	Jean	Jean	NNP
07-kim-ai	286	4	-	-	HYPH
07-kim-ai	286	5	François	François	NNP
07-kim-ai	286	6	,	,	,
07-kim-ai	286	7	Azim	Azim	NNP
07-kim-ai	286	8	Shariff	Shariff	NNP
07-kim-ai	286	9	,	,	,
07-kim-ai	286	10	and	and	CC
07-kim-ai	286	11	Iyad	Iyad	NNP
07-kim-ai	286	12	Rahwan	Rahwan	NNP
07-kim-ai	286	13	.	.	.
07-kim-ai	287	1	2016	2016	CD
07-kim-ai	287	2	.	.	.
07-kim-ai	288	1	“	"	``
07-kim-ai	288	2	The	the	DT
07-kim-ai	288	3	Social	Social	NNP
07-kim-ai	288	4	Dilemma	Dilemma	NNP
07-kim-ai	288	5	of	of	IN
07-kim-ai	288	6	Au-	Au-	NNP
07-kim-ai	288	7	tonomous	tonomous	JJ
07-kim-ai	288	8	Vehicles	vehicle	NNS
07-kim-ai	288	9	.	.	.
07-kim-ai	288	10	”	"	''
07-kim-ai	288	11	Science	science	NN
07-kim-ai	288	12	352	352	CD
07-kim-ai	288	13	(	(	-LRB-
07-kim-ai	288	14	6293	6293	CD
07-kim-ai	288	15	)	)	-RRB-
07-kim-ai	288	16	:	:	:
07-kim-ai	288	17	1573–76	1573–76	LS
07-kim-ai	288	18	.	.	.
07-kim-ai	289	1	?	?	.
07-kim-ai	289	2	iiTb	iitb	UH
07-kim-ai	289	3	,	,	,
07-kim-ai	289	4	ff	ff	NNP
07-kim-ai	289	5	/	/	SYM
07-kim-ai	289	6	QBXQ`;fRyXRRkefb+B2	QBXQ`;fRyXRRkefb+B2	NNP
07-kim-ai	289	7	M+2X	m+2x	NN
07-kim-ai	289	8	�	�	NNP
07-kim-ai	289	9	�	�	NNP
07-kim-ai	289	10	7ke89	7ke89	CD
07-kim-ai	289	11	.	.	.
07-kim-ai	290	1	Clark	Clark	NNP
07-kim-ai	290	2	,	,	,
07-kim-ai	290	3	Liat	Liat	NNP
07-kim-ai	290	4	.	.	.
07-kim-ai	291	1	2012	2012	CD
07-kim-ai	291	2	.	.	.
07-kim-ai	292	1	“	"	``
07-kim-ai	292	2	Google	Google	NNP
07-kim-ai	292	3	’s	’s	NNP
07-kim-ai	292	4	Artificial	Artificial	NNP
07-kim-ai	292	5	Brain	Brain	NNP
07-kim-ai	292	6	Learns	learn	VBZ
07-kim-ai	292	7	to	to	TO
07-kim-ai	292	8	Find	find	VB
07-kim-ai	292	9	Cat	cat	NN
07-kim-ai	292	10	Videos	video	NNS
07-kim-ai	292	11	.	.	.
07-kim-ai	292	12	”	"	''
07-kim-ai	292	13	Wired	wire	VBN
07-kim-ai	292	14	,	,	,
07-kim-ai	292	15	June	June	NNP
07-kim-ai	292	16	26	26	CD
07-kim-ai	292	17	,	,	,
07-kim-ai	292	18	2012	2012	CD
07-kim-ai	292	19	.	.	.
07-kim-ai	293	1	?	?	.
07-kim-ai	293	2	iiTb	iitb	UH
07-kim-ai	293	3	,	,	,
07-kim-ai	293	4	ffrrrXrB`2	ffrrrXrB`2	NNP
07-kim-ai	293	5	/	/	SYM
07-kim-ai	293	6	X+QKfkyRkfyef;QQ;H2@t@M2m`	X+QKfkyRkfyef;QQ;H2@t@M2m`	NNP
07-kim-ai	293	7	�	�	NNP
07-kim-ai	293	8	H@M2irQ`Ff	H@M2irQ`Ff	NNP
07-kim-ai	293	9	.	.	.
07-kim-ai	294	1	Conitzer	Conitzer	NNP
07-kim-ai	294	2	,	,	,
07-kim-ai	294	3	Vincent	Vincent	NNP
07-kim-ai	294	4	,	,	,
07-kim-ai	294	5	Walter	Walter	NNP
07-kim-ai	294	6	Sinnott	Sinnott	NNP
07-kim-ai	294	7	-	-	HYPH
07-kim-ai	294	8	Armstrong	Armstrong	NNP
07-kim-ai	294	9	,	,	,
07-kim-ai	294	10	Jana	Jana	NNP
07-kim-ai	294	11	Schaich	Schaich	NNP
07-kim-ai	294	12	Borg	Borg	NNP
07-kim-ai	294	13	,	,	,
07-kim-ai	294	14	Yuan	Yuan	NNP
07-kim-ai	294	15	Deng	Deng	NNP
07-kim-ai	294	16	,	,	,
07-kim-ai	294	17	and	and	CC
07-kim-ai	294	18	Max	Max	NNP
07-kim-ai	294	19	Kramer	Kramer	NNP
07-kim-ai	294	20	.	.	.
07-kim-ai	295	1	2017	2017	CD
07-kim-ai	295	2	.	.	.
07-kim-ai	296	1	“	"	``
07-kim-ai	296	2	Moral	moral	JJ
07-kim-ai	296	3	Decision	decision	NN
07-kim-ai	296	4	Making	make	VBG
07-kim-ai	296	5	Frameworks	Frameworks	NNP
07-kim-ai	296	6	for	for	IN
07-kim-ai	296	7	Artificial	Artificial	NNP
07-kim-ai	296	8	Intelligence	Intelligence	NNP
07-kim-ai	296	9	.	.	.
07-kim-ai	296	10	”	"	''
07-kim-ai	296	11	In	in	IN
07-kim-ai	296	12	Proceedingsofthe	Proceedingsofthe	NNP
07-kim-ai	296	13	Thirty	Thirty	NNP
07-kim-ai	296	14	-	-	HYPH
07-kim-ai	296	15	First	First	NNP
07-kim-ai	296	16	AAAI	AAAI	NNP
07-kim-ai	296	17	Conference	Conference	NNP
07-kim-ai	296	18	on	on	IN
07-kim-ai	296	19	Artificial	Artificial	NNP
07-kim-ai	296	20	Intelligence	Intelligence	NNP
07-kim-ai	296	21	,	,	,
07-kim-ai	296	22	4831–4835	4831–4835	CD
07-kim-ai	296	23	.	.	.
07-kim-ai	297	1	AAAI’17	aaai’17	UH
07-kim-ai	297	2	.	.	.
07-kim-ai	298	1	San	San	NNP
07-kim-ai	298	2	Fran-	Fran-	NNP
07-kim-ai	298	3	cisco	cisco	NN
07-kim-ai	298	4	,	,	,
07-kim-ai	298	5	California	California	NNP
07-kim-ai	298	6	,	,	,
07-kim-ai	298	7	USA	USA	NNP
07-kim-ai	298	8	:	:	:
07-kim-ai	298	9	AAAI	AAAI	NNP
07-kim-ai	298	10	Press	Press	NNP
07-kim-ai	298	11	.	.	.
07-kim-ai	299	1	Courtland	Courtland	NNP
07-kim-ai	299	2	,	,	,
07-kim-ai	299	3	Rachel	Rachel	NNP
07-kim-ai	299	4	.	.	.
07-kim-ai	300	1	2018	2018	CD
07-kim-ai	300	2	.	.	.
07-kim-ai	301	1	“	"	``
07-kim-ai	301	2	Bias	Bias	NNP
07-kim-ai	301	3	Detectives	detective	NNS
07-kim-ai	301	4	:	:	:
07-kim-ai	301	5	The	the	DT
07-kim-ai	301	6	Researchers	researcher	NNS
07-kim-ai	301	7	Striving	strive	VBG
07-kim-ai	301	8	to	to	TO
07-kim-ai	301	9	Make	make	VB
07-kim-ai	301	10	Algorithms	Algorithms	NNP
07-kim-ai	301	11	Fair	Fair	NNP
07-kim-ai	301	12	.	.	.
07-kim-ai	301	13	”	"	''
07-kim-ai	301	14	Nature	nature	NN
07-kim-ai	301	15	558	558	CD
07-kim-ai	301	16	(	(	-LRB-
07-kim-ai	301	17	7710	7710	CD
07-kim-ai	301	18	)	)	-RRB-
07-kim-ai	301	19	:	:	:
07-kim-ai	301	20	357–60	357–60	CD
07-kim-ai	301	21	.	.	.
07-kim-ai	302	1	?	?	.
07-kim-ai	302	2	iiTb	iitb	UH
07-kim-ai	302	3	,	,	,
07-kim-ai	302	4	ff	ff	VB
07-kim-ai	302	5	/	/	SYM
07-kim-ai	302	6	QBXQ`;fRyXRyj3f/9R83e@yR3@y89eN@j	qbxq`;fryxryj3f/9r83e@yr3@y89en@j	JJ
07-kim-ai	302	7	.	.	.
07-kim-ai	302	8	Cushman	Cushman	NNP
07-kim-ai	302	9	,	,	,
07-kim-ai	302	10	Fiery	Fiery	NNP
07-kim-ai	302	11	,	,	,
07-kim-ai	302	12	Liane	Liane	NNP
07-kim-ai	302	13	Young	Young	NNP
07-kim-ai	302	14	,	,	,
07-kim-ai	302	15	and	and	CC
07-kim-ai	302	16	Marc	Marc	NNP
07-kim-ai	302	17	Hauser	Hauser	NNP
07-kim-ai	302	18	.	.	.
07-kim-ai	303	1	2006	2006	CD
07-kim-ai	303	2	.	.	.
07-kim-ai	304	1	“	"	``
07-kim-ai	304	2	The	the	DT
07-kim-ai	304	3	Role	Role	NNP
07-kim-ai	304	4	of	of	IN
07-kim-ai	304	5	Conscious	Conscious	NNP
07-kim-ai	304	6	Reasoning	Reasoning	NNP
07-kim-ai	304	7	and	and	CC
07-kim-ai	304	8	Intuition	Intuition	NNP
07-kim-ai	304	9	in	in	IN
07-kim-ai	304	10	Moral	Moral	NNP
07-kim-ai	304	11	Judgment	Judgment	NNP
07-kim-ai	304	12	:	:	:
07-kim-ai	304	13	Testing	test	VBG
07-kim-ai	304	14	Three	three	CD
07-kim-ai	304	15	Principles	principle	NNS
07-kim-ai	304	16	of	of	IN
07-kim-ai	304	17	Harm	harm	NN
07-kim-ai	304	18	.	.	.
07-kim-ai	304	19	”	"	''
07-kim-ai	304	20	Psychological	Psychological	NNP
07-kim-ai	304	21	Science	Science	NNP
07-kim-ai	304	22	17	17	CD
07-kim-ai	304	23	(	(	-LRB-
07-kim-ai	304	24	12	12	CD
07-kim-ai	304	25	)	)	-RRB-
07-kim-ai	304	26	:	:	:
07-kim-ai	304	27	1082–89	1082–89	LS
07-kim-ai	304	28	.	.	.
07-kim-ai	305	1	Davis	Davis	NNP
07-kim-ai	305	2	,	,	,
07-kim-ai	305	3	Daniel	Daniel	NNP
07-kim-ai	305	4	L.	L.	NNP
07-kim-ai	305	5	2007	2007	CD
07-kim-ai	305	6	.	.	.
07-kim-ai	306	1	“	"	``
07-kim-ai	306	2	Who	who	WP
07-kim-ai	306	3	Decides	decide	VBZ
07-kim-ai	306	4	:	:	:
07-kim-ai	306	5	Man	man	NN
07-kim-ai	306	6	or	or	CC
07-kim-ai	306	7	Machine	machine	NN
07-kim-ai	306	8	?	?	.
07-kim-ai	306	9	”	"	''
07-kim-ai	306	10	Armed	Armed	NNP
07-kim-ai	306	11	Forces	Forces	NNPS
07-kim-ai	306	12	Journal	Journal	NNP
07-kim-ai	306	13	,	,	,
07-kim-ai	306	14	November	November	NNP
07-kim-ai	306	15	.	.	.
07-kim-ai	307	1	?	?	.
07-kim-ai	307	2	iiT	iit	JJ
07-kim-ai	307	3	,	,	,
07-kim-ai	307	4	ff	ff	NNP
07-kim-ai	307	5	�	�	NNP
07-kim-ai	307	6	`K2/7Q`+2bDQm`M	`K2/7Q`+2bDQm`M	NNP
07-kim-ai	307	7	�	�	:
07-kim-ai	307	8	HX+QKfr?Q@/2+B/2b@K	HX+QKfr?Q@/2+B/2b@K	NNS
07-kim-ai	307	9	�	�	NNS
07-kim-ai	307	10	M@Q`@K	M@Q`@K	NNS
07-kim-ai	307	11	�	�	NNP
07-kim-ai	307	12	+?BM2f	+?BM2f	NNP
07-kim-ai	307	13	.	.	.
07-kim-ai	308	1	Davis	Davis	NNP
07-kim-ai	308	2	,	,	,
07-kim-ai	308	3	Hannah	Hannah	NNP
07-kim-ai	308	4	.	.	.
07-kim-ai	309	1	2020	2020	CD
07-kim-ai	309	2	.	.	.
07-kim-ai	310	1	“	"	``
07-kim-ai	310	2	A	a	DT
07-kim-ai	310	3	Dataset	dataset	NN
07-kim-ai	310	4	Is	be	VBZ
07-kim-ai	310	5	a	a	DT
07-kim-ai	310	6	Worldview	worldview	NN
07-kim-ai	310	7	.	.	.
07-kim-ai	310	8	”	"	''
07-kim-ai	310	9	Towards	towards	IN
07-kim-ai	310	10	Data	Data	NNP
07-kim-ai	310	11	Science	Science	NNP
07-kim-ai	310	12	.	.	.
07-kim-ai	311	1	March	March	NNP
07-kim-ai	311	2	5	5	CD
07-kim-ai	311	3	,	,	,
07-kim-ai	311	4	2020	2020	CD
07-kim-ai	311	5	.	.	.
07-kim-ai	312	1	?	?	.
07-kim-ai	312	2	iiT	iiT	NNP
07-kim-ai	312	3	b	b	NN
07-kim-ai	312	4	,	,	,
07-kim-ai	312	5	ffiQr	ffiqr	ADD
07-kim-ai	312	6	�	�	NNP
07-kim-ai	312	7	`/b/	`/b/	NNP
07-kim-ai	312	8	�	�	NNP
07-kim-ai	312	9	i	i	PRP
07-kim-ai	312	10	�	�	:
07-kim-ai	312	11	b+B2M+2X+QKf	b+B2M+2X+QKf	NNS
07-kim-ai	312	12	�	�	NNP
07-kim-ai	312	13	@/	@/	FW
07-kim-ai	312	14	�	�	NNP
07-kim-ai	312	15	i	i	PRP
07-kim-ai	312	16	�	�	.
07-kim-ai	312	17	b2i@Bb@	b2i@Bb@	NNS
07-kim-ai	312	18	�	�	JJ
07-kim-ai	312	19	@rQ`H	@rQ`H	.
07-kim-ai	312	20	/	/	SYM
07-kim-ai	312	21	pB2r@8jk3kRe//99/.	pB2r@8jk3kRe//99/.	NNP
07-kim-ai	313	1	https://plato.stanford.edu/archives/win2016/entries/ethics-deontological/	https://plato.stanford.edu/archives/win2016/entries/ethics-deontological/	ADD
07-kim-ai	313	2	https://plato.stanford.edu/archives/win2016/entries/ethics-deontological/	https://plato.stanford.edu/archives/win2016/entries/ethics-deontological/	ADD
07-kim-ai	313	3	https://doi.org/10.1145/3306618.3314243	https://doi.org/10.1145/3306618.3314243	ADD
07-kim-ai	313	4	https://doi.org/10.1145/3306618.3314243	https://doi.org/10.1145/3306618.3314243	NNP
07-kim-ai	313	5	https://2020.code4lib.org/talks/AI-is-such-a-tool-Keeping-your-machine-learning-outputs-in-check	https://2020.code4lib.org/talks/AI-is-such-a-tool-Keeping-your-machine-learning-outputs-in-check	NNP
07-kim-ai	313	6	https://2020.code4lib.org/talks/AI-is-such-a-tool-Keeping-your-machine-learning-outputs-in-check	https://2020.code4lib.org/talks/ai-is-such-a-tool-keeping-your-machine-learning-outputs-in-check	CD
07-kim-ai	313	7	https://acrl.ala.org/techconnect/post/reflections-on-code4lib-2018/	https://acrl.ala.org/techconnect/post/reflections-on-code4lib-2018/	NNP
07-kim-ai	313	8	https://acrl.ala.org/techconnect/post/reflections-on-code4lib-2018/	https://acrl.ala.org/techconnect/post/reflections-on-code4lib-2018/	NNP
07-kim-ai	313	9	https://doi.org/10.1126/science.aaf2654	https://doi.org/10.1126/science.aaf2654	NNP
07-kim-ai	313	10	https://doi.org/10.1126/science.aaf2654	https://doi.org/10.1126/science.aaf2654	NNP
07-kim-ai	313	11	https://www.wired.com/2012/06/google-x-neural-network/	https://www.wired.com/2012/06/google-x-neural-network/	CD
07-kim-ai	313	12	https://doi.org/10.1038/d41586-018-05469-3	https://doi.org/10.1038/d41586-018-05469-3	CD
07-kim-ai	313	13	http://armedforcesjournal.com/who-decides-man-or-machine/	http://armedforcesjournal.com/who-decides-man-or-machine/	CD
07-kim-ai	313	14	https://towardsdatascience.com/a-dataset-is-a-worldview-5328216dd44d	https://towardsdatascience.com/a-dataset-is-a-worldview-5328216dd44d	NNS
07-kim-ai	313	15	https://towardsdatascience.com/a-dataset-is-a-worldview-5328216dd44d	https://towardsdatascience.com/a-dataset-is-a-worldview-5328216dd44d	VBZ
07-kim-ai	313	16	84	84	CD
07-kim-ai	313	17	Machine	Machine	NNP
07-kim-ai	313	18	Learning	Learning	NNP
07-kim-ai	313	19	,	,	,
07-kim-ai	313	20	Libraries	Libraries	NNPS
07-kim-ai	313	21	,	,	,
07-kim-ai	313	22	and	and	CC
07-kim-ai	313	23	Cross	Cross	NNP
07-kim-ai	313	24	-	-	NNP
07-kim-ai	313	25	Disciplinary	Disciplinary	NNP
07-kim-ai	313	26	ResearchǔChapter	ResearchǔChapter	NNP
07-kim-ai	313	27	7	7	CD
07-kim-ai	313	28	Foot	Foot	NNP
07-kim-ai	313	29	,	,	,
07-kim-ai	313	30	Philippa	Philippa	NNP
07-kim-ai	313	31	.	.	.
07-kim-ai	314	1	1967	1967	CD
07-kim-ai	314	2	.	.	.
07-kim-ai	315	1	“	"	``
07-kim-ai	315	2	The	the	DT
07-kim-ai	315	3	Problem	problem	NN
07-kim-ai	315	4	of	of	IN
07-kim-ai	315	5	Abortion	abortion	NN
07-kim-ai	315	6	and	and	CC
07-kim-ai	315	7	the	the	DT
07-kim-ai	315	8	Doctrine	Doctrine	NNP
07-kim-ai	315	9	of	of	IN
07-kim-ai	315	10	Double	Double	NNP
07-kim-ai	315	11	Effect	Effect	NNP
07-kim-ai	315	12	.	.	.
07-kim-ai	315	13	”	"	''
07-kim-ai	315	14	Oxford	Oxford	NNP
07-kim-ai	315	15	Review	Review	NNP
07-kim-ai	315	16	5	5	CD
07-kim-ai	315	17	:	:	:
07-kim-ai	315	18	5–15	5–15	CD
07-kim-ai	315	19	.	.	.
07-kim-ai	316	1	Heaven	Heaven	NNP
07-kim-ai	316	2	,	,	,
07-kim-ai	316	3	Will	Will	NNP
07-kim-ai	316	4	Douglas	Douglas	NNP
07-kim-ai	316	5	.	.	.
07-kim-ai	317	1	2020	2020	CD
07-kim-ai	317	2	.	.	.
07-kim-ai	318	1	“	"	``
07-kim-ai	318	2	DeepMind	DeepMind	NNP
07-kim-ai	318	3	’s	’s	,
07-kim-ai	318	4	AI	ai	NN
07-kim-ai	318	5	Can	Can	MD
07-kim-ai	318	6	Now	now	RB
07-kim-ai	318	7	Play	play	VB
07-kim-ai	318	8	All	all	DT
07-kim-ai	318	9	57	57	CD
07-kim-ai	318	10	Atari	Atari	NNP
07-kim-ai	318	11	Games	Games	NNPS
07-kim-ai	318	12	—	—	:
07-kim-ai	318	13	but	but	CC
07-kim-ai	318	14	It	-PRON-	PRP
07-kim-ai	318	15	’s	’	VBZ
07-kim-ai	318	16	Still	still	RB
07-kim-ai	318	17	Not	not	RB
07-kim-ai	318	18	Versatile	versatile	JJ
07-kim-ai	318	19	Enough	enough	RB
07-kim-ai	318	20	.	.	.
07-kim-ai	318	21	”	"	''
07-kim-ai	318	22	MIT	MIT	NNP
07-kim-ai	318	23	Technology	Technology	NNP
07-kim-ai	318	24	Review	Review	NNP
07-kim-ai	318	25	,	,	,
07-kim-ai	318	26	April	April	NNP
07-kim-ai	318	27	1	1	CD
07-kim-ai	318	28	,	,	,
07-kim-ai	318	29	2020	2020	CD
07-kim-ai	318	30	.	.	.
07-kim-ai	319	1	?	?	.
07-kim-ai	319	2	iiTb	iitb	UH
07-kim-ai	319	3	,	,	,
07-kim-ai	319	4	ffrrrXi2+?MQ	ffrrrxi2+?mq	JJ
07-kim-ai	319	5	HQ;v`2pB2rX+QKfkykyfy9fyRfNd9NNd	HQ;v`2pB2rX+QKfkykyfy9fyRfNd9NNd	NNP
07-kim-ai	319	6	.	.	.
07-kim-ai	320	1	International	International	NNP
07-kim-ai	320	2	Committee	Committee	NNP
07-kim-ai	320	3	of	of	IN
07-kim-ai	320	4	the	the	DT
07-kim-ai	320	5	Red	Red	NNP
07-kim-ai	320	6	Cross	Cross	NNP
07-kim-ai	320	7	.	.	.
07-kim-ai	321	1	2015	2015	CD
07-kim-ai	321	2	.	.	.
07-kim-ai	322	1	“	"	``
07-kim-ai	322	2	What	what	WP
07-kim-ai	322	3	Are	be	VBP
07-kim-ai	322	4	Jus	Jus	NNP
07-kim-ai	322	5	Ad	Ad	NNP
07-kim-ai	322	6	Bellum	Bellum	NNP
07-kim-ai	322	7	and	and	CC
07-kim-ai	322	8	Jus	Jus	NNP
07-kim-ai	322	9	in	in	IN
07-kim-ai	322	10	Bello	Bello	NNP
07-kim-ai	322	11	?	?	.
07-kim-ai	322	12	”	"	''
07-kim-ai	322	13	January	January	NNP
07-kim-ai	322	14	22	22	CD
07-kim-ai	322	15	,	,	,
07-kim-ai	322	16	2015	2015	CD
07-kim-ai	322	17	.	.	.
07-kim-ai	323	1	?	?	.
07-kim-ai	323	2	iiTb	iitb	UH
07-kim-ai	323	3	,	,	,
07-kim-ai	323	4	ffrrrXB+`+XQ`;f2Mf	ffrrrxb+`+xq`;f2mf	NN
07-kim-ai	323	5	/	/	SYM
07-kim-ai	323	6	Q+mK2Mifr?	q+mk2mifr?	ADD
07-kim-ai	323	7	�	�	VBZ
07-kim-ai	323	8	i@	i@	FW
07-kim-ai	323	9	�	�	NNS
07-kim-ai	323	10	`2@Dmb@	`2@Dmb@	:
07-kim-ai	323	11	�	�	NNS
07-kim-ai	323	12	/@#2H	/@#2H	,
07-kim-ai	323	13	HmK@	HmK@	NNP
07-kim-ai	323	14	�	�	NNP
07-kim-ai	323	15	M/@Dmb@#2HHQ@y	m/@dmb@#2hhq@y	NN
07-kim-ai	323	16	.	.	.
07-kim-ai	323	17	Kahn	Kahn	NNP
07-kim-ai	323	18	,	,	,
07-kim-ai	323	19	Leonard	Leonard	NNP
07-kim-ai	323	20	.	.	.
07-kim-ai	324	1	2012	2012	CD
07-kim-ai	324	2	.	.	.
07-kim-ai	325	1	“	"	``
07-kim-ai	325	2	Military	military	JJ
07-kim-ai	325	3	Robots	Robots	NNP
07-kim-ai	325	4	and	and	CC
07-kim-ai	325	5	The	the	DT
07-kim-ai	325	6	Likelihood	Likelihood	NNP
07-kim-ai	325	7	of	of	IN
07-kim-ai	325	8	Armed	Armed	NNP
07-kim-ai	325	9	Combat	Combat	NNP
07-kim-ai	325	10	.	.	.
07-kim-ai	325	11	”	"	''
07-kim-ai	325	12	In	in	IN
07-kim-ai	325	13	Robot	Robot	NNP
07-kim-ai	325	14	Ethics	Ethics	NNP
07-kim-ai	325	15	:	:	:
07-kim-ai	325	16	The	the	DT
07-kim-ai	325	17	Ethical	Ethical	NNP
07-kim-ai	325	18	and	and	CC
07-kim-ai	325	19	Social	Social	NNP
07-kim-ai	325	20	Implications	Implications	NNPS
07-kim-ai	325	21	of	of	IN
07-kim-ai	325	22	Robotics	Robotics	NNPS
07-kim-ai	325	23	,	,	,
07-kim-ai	325	24	edited	edit	VBN
07-kim-ai	325	25	by	by	IN
07-kim-ai	325	26	Patrick	Patrick	NNP
07-kim-ai	325	27	Lin	Lin	NNP
07-kim-ai	325	28	,	,	,
07-kim-ai	325	29	Keith	Keith	NNP
07-kim-ai	325	30	Abney	Abney	NNP
07-kim-ai	325	31	,	,	,
07-kim-ai	325	32	and	and	CC
07-kim-ai	325	33	George	George	NNP
07-kim-ai	325	34	A.	A.	NNP
07-kim-ai	325	35	Bekey	Bekey	NNP
07-kim-ai	325	36	,	,	,
07-kim-ai	325	37	274–92	274–92	CD
07-kim-ai	325	38	.	.	.
07-kim-ai	326	1	Intelligent	intelligent	JJ
07-kim-ai	326	2	Robotics	Robotics	NNPS
07-kim-ai	326	3	and	and	CC
07-kim-ai	326	4	Autonomous	Autonomous	NNP
07-kim-ai	326	5	Agents	Agents	NNPS
07-kim-ai	326	6	.	.	.
07-kim-ai	327	1	Cambridge	Cambridge	NNP
07-kim-ai	327	2	,	,	,
07-kim-ai	327	3	Mass.	Massachusetts	NNP
07-kim-ai	328	1	:	:	:
07-kim-ai	328	2	MIT	MIT	NNP
07-kim-ai	328	3	Press	Press	NNP
07-kim-ai	328	4	.	.	.
07-kim-ai	329	1	Knight	Knight	NNP
07-kim-ai	329	2	,	,	,
07-kim-ai	329	3	Will	Will	NNP
07-kim-ai	329	4	.	.	.
07-kim-ai	330	1	2017	2017	CD
07-kim-ai	330	2	.	.	.
07-kim-ai	331	1	“	"	``
07-kim-ai	331	2	The	the	DT
07-kim-ai	331	3	Dark	Dark	NNP
07-kim-ai	331	4	Secret	Secret	NNP
07-kim-ai	331	5	at	at	IN
07-kim-ai	331	6	the	the	DT
07-kim-ai	331	7	Heart	heart	NN
07-kim-ai	331	8	of	of	IN
07-kim-ai	331	9	AI	AI	NNP
07-kim-ai	331	10	.	.	.
07-kim-ai	331	11	”	"	''
07-kim-ai	331	12	MIT	MIT	NNP
07-kim-ai	331	13	Technology	Technology	NNP
07-kim-ai	331	14	Review	Review	NNP
07-kim-ai	331	15	,	,	,
07-kim-ai	331	16	April	April	NNP
07-kim-ai	331	17	11	11	CD
07-kim-ai	331	18	,	,	,
07-kim-ai	331	19	2017	2017	CD
07-kim-ai	331	20	.	.	.
07-kim-ai	332	1	?	?	.
07-kim-ai	332	2	iiTb	iitb	UH
07-kim-ai	332	3	,	,	,
07-kim-ai	332	4	ffrrrXi2+?MQHQ;v`2pB2rX+QKfkyRdfy9fRRf8RRj	ffrrrxi2+?mqhq;v`2pb2rx+qkfkyrdfy9frrf8rrj	NN
07-kim-ai	332	5	.	.	.
07-kim-ai	333	1	.	.	.
07-kim-ai	334	1	2019a	2019a	CD
07-kim-ai	334	2	.	.	.
07-kim-ai	335	1	“	"	``
07-kim-ai	335	2	Two	two	CD
07-kim-ai	335	3	Rival	rival	JJ
07-kim-ai	335	4	AI	AI	NNP
07-kim-ai	335	5	Approaches	approach	NNS
07-kim-ai	335	6	Combine	combine	VBP
07-kim-ai	335	7	to	to	TO
07-kim-ai	335	8	Let	let	VB
07-kim-ai	335	9	Machines	machine	NNS
07-kim-ai	335	10	Learn	learn	VB
07-kim-ai	335	11	about	about	IN
07-kim-ai	335	12	the	the	DT
07-kim-ai	335	13	World	World	NNP
07-kim-ai	335	14	like	like	IN
07-kim-ai	335	15	a	a	DT
07-kim-ai	335	16	Child	child	NN
07-kim-ai	335	17	.	.	.
07-kim-ai	335	18	”	"	''
07-kim-ai	335	19	MIT	MIT	NNP
07-kim-ai	335	20	Technology	Technology	NNP
07-kim-ai	335	21	Review	Review	NNP
07-kim-ai	335	22	,	,	,
07-kim-ai	335	23	April	April	NNP
07-kim-ai	335	24	8	8	CD
07-kim-ai	335	25	,	,	,
07-kim-ai	335	26	2019	2019	CD
07-kim-ai	335	27	.	.	.
07-kim-ai	336	1	?	?	.
07-kim-ai	336	2	iiTb	iitb	UH
07-kim-ai	336	3	,	,	,
07-kim-ai	336	4	ffrrrXi2+?MQHQ;v	ffrrrXi2+?MQHQ;v	NNP
07-kim-ai	336	5	`	`	''
07-kim-ai	336	6	2pB2rX+QKfkyRNfy9fy3fRyjkkj	2pb2rx+qkfkyrnfy9fy3fryjkkj	CD
07-kim-ai	336	7	.	.	.
07-kim-ai	337	1	.	.	.
07-kim-ai	338	1	2019b	2019b	CD
07-kim-ai	338	2	.	.	.
07-kim-ai	339	1	“	"	``
07-kim-ai	339	2	AI	AI	NNP
07-kim-ai	339	3	Is	be	VBZ
07-kim-ai	339	4	Biased	bias	VBN
07-kim-ai	339	5	.	.	.
07-kim-ai	340	1	Here	here	RB
07-kim-ai	340	2	’s	’	VBZ
07-kim-ai	340	3	How	how	WRB
07-kim-ai	340	4	Scientists	scientist	NNS
07-kim-ai	340	5	Are	be	VBP
07-kim-ai	340	6	Trying	try	VBG
07-kim-ai	340	7	to	to	TO
07-kim-ai	340	8	Fix	fix	VB
07-kim-ai	340	9	It	-PRON-	PRP
07-kim-ai	340	10	.	.	.
07-kim-ai	340	11	”	"	''
07-kim-ai	340	12	Wired	wire	VBN
07-kim-ai	340	13	,	,	,
07-kim-ai	340	14	De-	de-	UH
07-kim-ai	340	15	cember	cember	NN
07-kim-ai	340	16	19	19	CD
07-kim-ai	340	17	,	,	,
07-kim-ai	340	18	2019	2019	CD
07-kim-ai	340	19	.	.	.
07-kim-ai	341	1	?	?	.
07-kim-ai	341	2	iiTb	iitb	UH
07-kim-ai	341	3	,	,	,
07-kim-ai	341	4	ffrrrXrB`2	ffrrrxrb`2	NN
07-kim-ai	341	5	/	/	SYM
07-kim-ai	341	6	X+QKfbiQ`vf	x+qkfbiq`vf	NN
07-kim-ai	341	7	�	�	ADD
07-kim-ai	341	8	B@#B	b@#b	CD
07-kim-ai	341	9	�	�	NN
07-kim-ai	341	10	b2/@?Qr@b+B2MiBbib	b2/@?qr@b+b2mibbib	CD
07-kim-ai	341	11	@i`vBM;@7Btf	@i`vBM;@7Btf	NFP
07-kim-ai	341	12	.	.	.
07-kim-ai	342	1	Koch	Koch	NNP
07-kim-ai	342	2	,	,	,
07-kim-ai	342	3	Christof	Christof	NNP
07-kim-ai	342	4	.	.	.
07-kim-ai	343	1	2016	2016	CD
07-kim-ai	343	2	.	.	.
07-kim-ai	344	1	“	"	``
07-kim-ai	344	2	How	how	WRB
07-kim-ai	344	3	the	the	DT
07-kim-ai	344	4	Computer	Computer	NNP
07-kim-ai	344	5	Beat	beat	VBP
07-kim-ai	344	6	the	the	DT
07-kim-ai	344	7	Go	Go	NNP
07-kim-ai	344	8	Master	Master	NNP
07-kim-ai	344	9	.	.	.
07-kim-ai	344	10	”	"	''
07-kim-ai	344	11	Scientific	scientific	JJ
07-kim-ai	344	12	American	American	NNP
07-kim-ai	344	13	.	.	.
07-kim-ai	345	1	March	March	NNP
07-kim-ai	345	2	19	19	CD
07-kim-ai	345	3	,	,	,
07-kim-ai	345	4	2016	2016	CD
07-kim-ai	345	5	.	.	.
07-kim-ai	346	1	?	?	.
07-kim-ai	346	2	iiTb	iitb	UH
07-kim-ai	346	3	,	,	,
07-kim-ai	346	4	ffrrrXb+B2MiB7B+	ffrrrXb+B2MiB7B+	NNP
07-kim-ai	346	5	�	�	.
07-kim-ai	346	6	K2`B+	K2`B+	NNS
07-kim-ai	346	7	�	�	NNP
07-kim-ai	346	8	MX+QKf	MX+QKf	NNS
07-kim-ai	346	9	�	�	NNS
07-kim-ai	346	10	`iB+H2f?Qr@i?2@+QKTmi2	`ib+h2f?qr@i?2@+qktmi2	NN
07-kim-ai	346	11	`	`	''
07-kim-ai	346	12	@#2	@#2	NFP
07-kim-ai	346	13	�	�	NNP
07-kim-ai	346	14	i@i?2@;Q@K	i@i?2@;Q@K	NNP
07-kim-ai	346	15	�	�	NNP
07-kim-ai	346	16	bi2`f	bi2`f	NNP
07-kim-ai	346	17	.	.	.
07-kim-ai	347	1	Lohr	Lohr	NNP
07-kim-ai	347	2	,	,	,
07-kim-ai	347	3	Steve	Steve	NNP
07-kim-ai	347	4	.	.	.
07-kim-ai	348	1	2018	2018	CD
07-kim-ai	348	2	.	.	.
07-kim-ai	349	1	“	"	``
07-kim-ai	349	2	Facial	facial	JJ
07-kim-ai	349	3	Recognition	Recognition	NNP
07-kim-ai	349	4	Is	be	VBZ
07-kim-ai	349	5	Accurate	accurate	JJ
07-kim-ai	349	6	,	,	,
07-kim-ai	349	7	If	if	IN
07-kim-ai	349	8	You	-PRON-	PRP
07-kim-ai	349	9	’re	be	VBP
07-kim-ai	349	10	a	a	DT
07-kim-ai	349	11	White	White	NNP
07-kim-ai	349	12	Guy	Guy	NNP
07-kim-ai	349	13	.	.	.
07-kim-ai	349	14	”	"	''
07-kim-ai	349	15	New	New	NNP
07-kim-ai	349	16	York	York	NNP
07-kim-ai	349	17	Times	Times	NNP
07-kim-ai	349	18	,	,	,
07-kim-ai	349	19	February	February	NNP
07-kim-ai	349	20	9	9	CD
07-kim-ai	349	21	,	,	,
07-kim-ai	349	22	2018	2018	CD
07-kim-ai	349	23	.	.	.
07-kim-ai	350	1	?	?	.
07-kim-ai	350	2	iiTb	iitb	UH
07-kim-ai	350	3	,	,	,
07-kim-ai	350	4	ffrrrXMviBK2bX+QKfkyR3fykfyNfi2+?MQHQ;vf7	ffrrrxmvibk2bx+qkfkyr3fykfynfi2+?mqhq;vf7	ADD
07-kim-ai	350	5	�	�	NNP
07-kim-ai	350	6	+B	+B	NNP
07-kim-ai	350	7	�	�	JJ
07-kim-ai	350	8	H	h	NN
07-kim-ai	350	9	@`2+Q;MBiBQM@`	@`2+Q;MBiBQM@`	:
07-kim-ai	350	10	�	�	ADD
07-kim-ai	350	11	+2@	+2@	LS
07-kim-ai	350	12	�	�	NNP
07-kim-ai	350	13	`iB7B+B	`iB7B+B	NNP
07-kim-ai	350	14	�	�	:
07-kim-ai	350	15	H@BMi2HHB;2M+2X?iKH	h@bmi2hhb;2m+2x?ikh	JJ
07-kim-ai	350	16	.	.	.
07-kim-ai	351	1	Mahlangu	Mahlangu	NNP
07-kim-ai	351	2	,	,	,
07-kim-ai	351	3	Isaac	Isaac	NNP
07-kim-ai	351	4	.	.	.
07-kim-ai	352	1	2019	2019	CD
07-kim-ai	352	2	.	.	.
07-kim-ai	353	1	“	"	``
07-kim-ai	353	2	Meet	Meet	NNP
07-kim-ai	353	3	Libby	Libby	NNP
07-kim-ai	353	4	-	-	,
07-kim-ai	353	5	the	the	DT
07-kim-ai	353	6	New	New	NNP
07-kim-ai	353	7	Robot	Robot	NNP
07-kim-ai	353	8	Library	Library	NNP
07-kim-ai	353	9	Assistant	Assistant	NNP
07-kim-ai	353	10	at	at	IN
07-kim-ai	353	11	the	the	DT
07-kim-ai	353	12	University	University	NNP
07-kim-ai	353	13	of	of	IN
07-kim-ai	353	14	Pretoria	Pretoria	NNP
07-kim-ai	353	15	’s	’s	NNP
07-kim-ai	353	16	Hatfield	Hatfield	NNP
07-kim-ai	353	17	Campus	Campus	NNP
07-kim-ai	353	18	.	.	.
07-kim-ai	353	19	”	"	''
07-kim-ai	353	20	SowetanLIVE	sowetanlive	CD
07-kim-ai	353	21	.	.	.
07-kim-ai	354	1	June	June	NNP
07-kim-ai	354	2	4	4	CD
07-kim-ai	354	3	,	,	,
07-kim-ai	354	4	2019	2019	CD
07-kim-ai	354	5	.	.	.
07-kim-ai	355	1	?	?	.
07-kim-ai	355	2	iiTb	iitb	UH
07-kim-ai	355	3	,	,	,
07-kim-ai	355	4	ffrrrXbQr2i	ffrrrXbQr2i	NNP
07-kim-ai	355	5	�	�	NNP
07-kim-ai	355	6	MHBp	MHBp	NNS
07-kim-ai	355	7	2X+QXx	2x+qxx	CD
07-kim-ai	355	8	�	�	NNS
07-kim-ai	355	9	fM2rbfbQmi	fM2rbfbQmi	NNS
07-kim-ai	355	10	?	?	.
07-kim-ai	356	1	@	@	NFP
07-kim-ai	356	2	�	�	NNP
07-kim-ai	356	3	7`B+	7`B+	NNS
07-kim-ai	356	4	�	�	JJ
07-kim-ai	356	5	fkyRN@ye@y9@K22i@HB##v@i?2@M2r@`Q#Qi@HB	fkyRN@ye@y9@K22i@HB##v@i?2@M2r@`Q#Qi@HB	.
07-kim-ai	356	6	#	#	$
07-kim-ai	356	7	`	`	''
07-kim-ai	356	8	�	�	NNP
07-kim-ai	356	9	`	`	''
07-kim-ai	356	10	v@	v@	NNP
07-kim-ai	356	11	�	�	NNP
07-kim-ai	356	12	bbBbi	bbbbi	NN
07-kim-ai	356	13	�	�	NNP
07-kim-ai	356	14	Mi@	mi@	NN
07-kim-ai	356	15	�	�	NNP
07-kim-ai	356	16	i@i?2@mMBp2`bBiv@Q7@T`2iQ`B	i@i?2@mMBp2`bBiv@Q7@T`2iQ`B	NNP
07-kim-ai	356	17	�	�	NNP
07-kim-ai	356	18	b@?	b@?	VBD
07-kim-ai	356	19	�	�	.
07-kim-ai	356	20	i7B2H/@+	i7b2h/@+	UH
07-kim-ai	356	21	�	�	NNP
07-kim-ai	356	22	KTmbf	KTmbf	NNP
07-kim-ai	356	23	.	.	.
07-kim-ai	357	1	Markoff	Markoff	NNP
07-kim-ai	357	2	,	,	,
07-kim-ai	357	3	John	John	NNP
07-kim-ai	357	4	.	.	.
07-kim-ai	358	1	2012	2012	CD
07-kim-ai	358	2	.	.	.
07-kim-ai	359	1	“	"	``
07-kim-ai	359	2	How	how	WRB
07-kim-ai	359	3	Many	many	JJ
07-kim-ai	359	4	Computers	computer	NNS
07-kim-ai	359	5	to	to	TO
07-kim-ai	359	6	Identify	identify	VB
07-kim-ai	359	7	a	a	DT
07-kim-ai	359	8	Cat	cat	NN
07-kim-ai	359	9	?	?	.
07-kim-ai	360	1	16,000	16,000	CD
07-kim-ai	360	2	.	.	.
07-kim-ai	360	3	”	"	''
07-kim-ai	360	4	New	New	NNP
07-kim-ai	360	5	York	York	NNP
07-kim-ai	360	6	Times	Times	NNP
07-kim-ai	360	7	,	,	,
07-kim-ai	360	8	June	June	NNP
07-kim-ai	360	9	25	25	CD
07-kim-ai	360	10	,	,	,
07-kim-ai	360	11	2012	2012	CD
07-kim-ai	360	12	.	.	.
07-kim-ai	361	1	Marr	Marr	NNP
07-kim-ai	361	2	,	,	,
07-kim-ai	361	3	Bernard	Bernard	NNP
07-kim-ai	361	4	.	.	.
07-kim-ai	362	1	2018	2018	CD
07-kim-ai	362	2	.	.	.
07-kim-ai	363	1	“	"	``
07-kim-ai	363	2	How	how	WRB
07-kim-ai	363	3	AI	AI	NNP
07-kim-ai	363	4	And	and	CC
07-kim-ai	363	5	Machine	Machine	NNP
07-kim-ai	363	6	Learning	Learning	NNP
07-kim-ai	363	7	Are	be	VBP
07-kim-ai	363	8	Transforming	transform	VBG
07-kim-ai	363	9	Law	Law	NNP
07-kim-ai	363	10	Firms	firm	NNS
07-kim-ai	363	11	And	and	CC
07-kim-ai	363	12	The	the	DT
07-kim-ai	363	13	Legal	Legal	NNP
07-kim-ai	363	14	Sector	Sector	NNP
07-kim-ai	363	15	.	.	.
07-kim-ai	363	16	”	"	''
07-kim-ai	363	17	Forbes	Forbes	NNP
07-kim-ai	363	18	,	,	,
07-kim-ai	363	19	May	May	NNP
07-kim-ai	363	20	23	23	CD
07-kim-ai	363	21	,	,	,
07-kim-ai	363	22	2018	2018	CD
07-kim-ai	363	23	.	.	.
07-kim-ai	364	1	?	?	.
07-kim-ai	364	2	iiTb	iitb	UH
07-kim-ai	364	3	,	,	,
07-kim-ai	364	4	ffrrrX7Q`#2bX+QKfbBi2bf#2`M	ffrrrx7q`#2bx+qkfbbi2bf#2`m	IN
07-kim-ai	364	5	�	�	NNP
07-kim-ai	364	6	`/K	`/K	NNP
07-kim-ai	364	7	�	�	NNP
07-kim-ai	364	8	`	`	''
07-kim-ai	364	9	`	`	''
07-kim-ai	364	10	fkyR3fy8fkjf?Qr@	fkyR3fy8fkjf?Qr@	NNPS
07-kim-ai	364	11	�	�	NNP
07-kim-ai	364	12	B@	B@	NNS
07-kim-ai	364	13	�	�	,
07-kim-ai	364	14	M/@K	m/@k	JJ
07-kim-ai	364	15	�	�	NNP
07-kim-ai	364	16	+?BM2@H2	+?BM2@H2	NNP
07-kim-ai	364	17	�	�	NNP
07-kim-ai	364	18	`MBM;@	`mbm;@	NN
07-kim-ai	364	19	�	�	NNS
07-kim-ai	364	20	`2@i`	`2@i`	JJS
07-kim-ai	364	21	�	�	NNS
07-kim-ai	364	22	Mb7Q`KBM;@H	Mb7Q`KBM;@H	NNPS
07-kim-ai	364	23	�	�	JJ
07-kim-ai	364	24	r@7	r@7	XX
07-kim-ai	364	25	B`Kb@	B`Kb@	NNP
07-kim-ai	364	26	�	�	NNP
07-kim-ai	364	27	M/@i?2@H2;	M/@i?2@H2;	NNPS
07-kim-ai	364	28	�	�	NNP
07-kim-ai	364	29	H@b2+iQ`f	H@b2+iQ`f	NNP
07-kim-ai	364	30	.	.	.
07-kim-ai	365	1	Pariser	Pariser	NNP
07-kim-ai	365	2	,	,	,
07-kim-ai	365	3	Eli	Eli	NNP
07-kim-ai	365	4	.	.	.
07-kim-ai	366	1	2011	2011	CD
07-kim-ai	366	2	.	.	.
07-kim-ai	367	1	TheFilterBubble	TheFilterBubble	NNP
07-kim-ai	367	2	:	:	:
07-kim-ai	367	3	HowtheNewPersonalizedWebIsChangingWhatWeRead	HowtheNewPersonalizedWebIsChangingWhatWeRead	NNS
07-kim-ai	367	4	and	and	CC
07-kim-ai	367	5	How	how	WRB
07-kim-ai	367	6	We	-PRON-	PRP
07-kim-ai	367	7	Think	think	VBP
07-kim-ai	367	8	.	.	.
07-kim-ai	368	1	New	New	NNP
07-kim-ai	368	2	York	York	NNP
07-kim-ai	368	3	:	:	:
07-kim-ai	368	4	Penguin	Penguin	NNP
07-kim-ai	368	5	Press	Press	NNP
07-kim-ai	368	6	.	.	.
07-kim-ai	369	1	Price	price	NN
07-kim-ai	369	2	,	,	,
07-kim-ai	369	3	Gary	Gary	NNP
07-kim-ai	369	4	.	.	.
07-kim-ai	370	1	2019	2019	CD
07-kim-ai	370	2	.	.	.
07-kim-ai	371	1	“	"	``
07-kim-ai	371	2	The	the	DT
07-kim-ai	371	3	Library	Library	NNP
07-kim-ai	371	4	of	of	IN
07-kim-ai	371	5	Congress	Congress	NNP
07-kim-ai	371	6	Posts	Posts	NNPS
07-kim-ai	371	7	Solicitation	Solicitation	NNP
07-kim-ai	371	8	For	for	IN
07-kim-ai	371	9	a	a	DT
07-kim-ai	371	10	Machine	Machine	NNP
07-kim-ai	371	11	Learning	Learning	NNP
07-kim-ai	371	12	/	/	SYM
07-kim-ai	371	13	Deep	Deep	NNP
07-kim-ai	371	14	Learning	Learning	NNP
07-kim-ai	371	15	Pilot	Pilot	NNP
07-kim-ai	371	16	Program	Program	NNP
07-kim-ai	371	17	to	to	TO
07-kim-ai	371	18	‘	'	``
07-kim-ai	371	19	Maximize	maximize	VB
07-kim-ai	371	20	the	the	DT
07-kim-ai	371	21	Use	Use	NNP
07-kim-ai	371	22	of	of	IN
07-kim-ai	371	23	Its	-PRON-	PRP$
07-kim-ai	371	24	Digital	Digital	NNP
07-kim-ai	371	25	Collection	Collection	NNP
07-kim-ai	371	26	.	.	.
07-kim-ai	371	27	’	'	''
07-kim-ai	371	28	”	"	''
07-kim-ai	371	29	LJ	LJ	NNP
07-kim-ai	371	30	InfoDOCKET	InfoDOCKET	NNP
07-kim-ai	371	31	.	.	.
07-kim-ai	372	1	June	June	NNP
07-kim-ai	372	2	13	13	CD
07-kim-ai	372	3	,	,	,
07-kim-ai	372	4	2019	2019	CD
07-kim-ai	372	5	.	.	.
07-kim-ai	373	1	?	?	.
07-kim-ai	373	2	iiTb	iitb	UH
07-kim-ai	373	3	,	,	,
07-kim-ai	373	4	ffrrrXBM7Q	ffrrrXBM7Q	''
07-kim-ai	373	5	/	/	SYM
07-kim-ai	373	6	Q+F2iX+QKfkyRNfyefRjfHB#`	Q+F2iX+QKfkyRNfyefRjfHB#`	NNP
07-kim-ai	373	7	�	�	NNP
07-kim-ai	373	8	`v@Q7@+QM	`v@q7@+qm	CD
07-kim-ai	373	9	;	;	:
07-kim-ai	373	10	`	`	''
07-kim-ai	373	11	2bb@TQbib@bQHB+Bi	2bb@tqbib@bqhb+bi	CD
07-kim-ai	373	12	�	�	NNS
07-kim-ai	373	13	iBQM@7Q`@	ibqm@7q`@	NN
07-kim-ai	373	14	�	�	.
07-kim-ai	373	15	@K	@k	NN
07-kim-ai	373	16	�	�	,
07-kim-ai	373	17	+?BM2@H2	+?BM2@H2	NNP
07-kim-ai	373	18	�	�	NNP
07-kim-ai	373	19	`MBM;@/22T@H2	`MBM;@/22T@H2	NNP
07-kim-ai	373	20	�	�	NNP
07-kim-ai	373	21	`MBM;@TBHQ	`MBM;@TBHQ	.
07-kim-ai	373	22	i@T`Q;`	i@t`q;`	FW
07-kim-ai	373	23	�	�	JJ
07-kim-ai	373	24	K@iQ@K	K@iQ@K	NNP
07-kim-ai	373	25	�	�	JJ
07-kim-ai	373	26	tBKBx2@i?2@mb2@Q7@Bib@/B;Bi	tBKBx2@i?2@mb2@Q7@Bib@/B;Bi	NNP
07-kim-ai	373	27	�	�	NNP
07-kim-ai	373	28	H@+QHH2+iBQM@HB#`	h@+qhh2+ibqm@hb#`	NN
07-kim-ai	373	29	�	�	NNP
07-kim-ai	373	30	`v@	`v@	NN
07-kim-ai	373	31	Bb@HQQFBM;@7Q`@`f	Bb@HQQFBM;@7Q`@`f	NNS
07-kim-ai	373	32	.	.	.
07-kim-ai	373	33	Rincon	Rincon	NNP
07-kim-ai	373	34	,	,	,
07-kim-ai	373	35	Lilian	Lilian	NNP
07-kim-ai	373	36	.	.	.
07-kim-ai	374	1	2019	2019	CD
07-kim-ai	374	2	.	.	.
07-kim-ai	375	1	“	"	``
07-kim-ai	375	2	Interpreter	Interpreter	NNP
07-kim-ai	375	3	Mode	Mode	NNP
07-kim-ai	375	4	Brings	bring	VBZ
07-kim-ai	375	5	Real	real	JJ
07-kim-ai	375	6	-	-	HYPH
07-kim-ai	375	7	Time	time	NN
07-kim-ai	375	8	Translation	translation	NN
07-kim-ai	375	9	to	to	IN
07-kim-ai	375	10	Your	-PRON-	PRP$
07-kim-ai	375	11	Phone	phone	NN
07-kim-ai	375	12	.	.	.
07-kim-ai	375	13	”	"	''
07-kim-ai	375	14	Google	Google	NNP
07-kim-ai	375	15	Blog	Blog	NNP
07-kim-ai	375	16	(	(	-LRB-
07-kim-ai	375	17	blog	blog	NN
07-kim-ai	375	18	)	)	-RRB-
07-kim-ai	375	19	.	.	.
07-kim-ai	376	1	December	December	NNP
07-kim-ai	376	2	12	12	CD
07-kim-ai	376	3	,	,	,
07-kim-ai	376	4	2019	2019	CD
07-kim-ai	376	5	.	.	.
07-kim-ai	377	1	?	?	.
07-kim-ai	377	2	iiTb	iitb	UH
07-kim-ai	377	3	,	,	,
07-kim-ai	377	4	ffrrrX#HQ;X;QQ;H2fT`Q	ffrrrX#HQ;X;QQ;H2fT`Q	NNP
07-kim-ai	377	5	/	/	SYM
07-kim-ai	377	6	m+ibf	m+ibf	NNP
07-kim-ai	377	7	�	�	NNP
07-kim-ai	377	8	bbBbi	bbBbi	NNP
07-kim-ai	377	9	�	�	NNP
07-kim-ai	377	10	https://www.technologyreview.com/2020/04/01/974997	https://www.technologyreview.com/2020/04/01/974997	:
07-kim-ai	377	11	https://www.technologyreview.com/2020/04/01/974997	https://www.technologyreview.com/2020/04/01/974997	:
07-kim-ai	377	12	https://www.icrc.org/en/document/what-are-jus-ad-bellum-and-jus-bello-0	https://www.icrc.org/en/document/what-are-jus-ad-bellum-and-jus-bello-0	NNP
07-kim-ai	377	13	https://www.icrc.org/en/document/what-are-jus-ad-bellum-and-jus-bello-0	https://www.icrc.org/en/document/what-are-jus-ad-bellum-and-jus-bello-0	NNP
07-kim-ai	377	14	https://www.technologyreview.com/2017/04/11/5113	https://www.technologyreview.com/2017/04/11/5113	NNP
07-kim-ai	377	15	https://www.technologyreview.com/2019/04/08/103223	https://www.technologyreview.com/2019/04/08/103223	ADD
07-kim-ai	377	16	https://www.technologyreview.com/2019/04/08/103223	https://www.technologyreview.com/2019/04/08/103223	.
07-kim-ai	377	17	https://www.wired.com/story/ai-biased-how-scientists-trying-fix/	https://www.wired.com/story/ai-biased-how-scientists-trying-fix/	NNP
07-kim-ai	377	18	https://www.wired.com/story/ai-biased-how-scientists-trying-fix/	https://www.wired.com/story/ai-biased-how-scientists-trying-fix/	NNP
07-kim-ai	377	19	https://www.scientificamerican.com/article/how-the-computer-beat-the-go-master/	https://www.scientificamerican.com/article/how-the-computer-beat-the-go-master/	NN
07-kim-ai	377	20	https://www.scientificamerican.com/article/how-the-computer-beat-the-go-master/	https://www.scientificamerican.com/article/how-the-computer-beat-the-go-master/	NN
07-kim-ai	377	21	https://www.nytimes.com/2018/02/09/technology/facial-recognition-race-artificial-intelligence.html	https://www.nytimes.com/2018/02/09/technology/facial-recognition-race-artificial-intelligence.html	NN
07-kim-ai	377	22	https://www.nytimes.com/2018/02/09/technology/facial-recognition-race-artificial-intelligence.html	https://www.nytimes.com/2018/02/09/technology/facial-recognition-race-artificial-intelligence.html	NN
07-kim-ai	377	23	https://www.sowetanlive.co.za/news/south-africa/2019-06-04-meet-libby-the-new-robot-library-assistant-at-the-university-of-pretorias-hatfield-campus/	https://www.sowetanlive.co.za/news/south-africa/2019-06-04-meet-libby-the-new-robot-library-assistant-at-the-university-of-pretorias-hatfield-campus/	NNP
07-kim-ai	377	24	https://www.sowetanlive.co.za/news/south-africa/2019-06-04-meet-libby-the-new-robot-library-assistant-at-the-university-of-pretorias-hatfield-campus/	https://www.sowetanlive.co.za/news/south-africa/2019-06-04-meet-libby-the-new-robot-library-assistant-at-the-university-of-pretorias-hatfield-campus/	NNP
07-kim-ai	377	25	https://www.sowetanlive.co.za/news/south-africa/2019-06-04-meet-libby-the-new-robot-library-assistant-at-the-university-of-pretorias-hatfield-campus/	https://www.sowetanlive.co.za/news/south-africa/2019-06-04-meet-libby-the-new-robot-library-assistant-at-the-university-of-pretorias-hatfield-campus/	.
07-kim-ai	377	26	https://www.forbes.com/sites/bernardmarr/2018/05/23/how-ai-and-machine-learning-are-transforming-law-firms-and-the-legal-sector/	https://www.forbes.com/sites/bernardmarr/2018/05/23/how-ai-and-machine-learning-are-transforming-law-firms-and-the-legal-sector/	ADD
07-kim-ai	377	27	https://www.forbes.com/sites/bernardmarr/2018/05/23/how-ai-and-machine-learning-are-transforming-law-firms-and-the-legal-sector/	https://www.forbes.com/sites/bernardmarr/2018/05/23/how-ai-and-machine-learning-are-transforming-law-firms-and-the-legal-sector/	ADD
07-kim-ai	377	28	https://www.forbes.com/sites/bernardmarr/2018/05/23/how-ai-and-machine-learning-are-transforming-law-firms-and-the-legal-sector/	https://www.forbes.com/sites/bernardmarr/2018/05/23/how-ai-and-machine-learning-are-transforming-law-firms-and-the-legal-sector/	NNP
07-kim-ai	377	29	https://www.infodocket.com/2019/06/13/library-of-congress-posts-solicitation-for-a-machine-learning-deep-learning-pilot-program-to-maximize-the-use-of-its-digital-collection-library-is-looking-for-r/	https://www.infodocket.com/2019/06/13/library-of-congress-posts-solicitation-for-a-machine-learning-deep-learning-pilot-program-to-maximize-the-use-of-its-digital-collection-library-is-looking-for-r/	NNP
07-kim-ai	377	30	https://www.infodocket.com/2019/06/13/library-of-congress-posts-solicitation-for-a-machine-learning-deep-learning-pilot-program-to-maximize-the-use-of-its-digital-collection-library-is-looking-for-r/	https://www.infodocket.com/2019/06/13/library-of-congress-posts-solicitation-for-a-machine-learning-deep-learning-pilot-program-to-maximize-the-use-of-its-digital-collection-library-is-looking-for-r/	JJ
07-kim-ai	377	31	https://www.infodocket.com/2019/06/13/library-of-congress-posts-solicitation-for-a-machine-learning-deep-learning-pilot-program-to-maximize-the-use-of-its-digital-collection-library-is-looking-for-r/	https://www.infodocket.com/2019/06/13/library-of-congress-posts-solicitation-for-a-machine-learning-deep-learning-pilot-program-to-maximize-the-use-of-its-digital-collection-library-is-looking-for-r/	JJ
07-kim-ai	377	32	https://www.infodocket.com/2019/06/13/library-of-congress-posts-solicitation-for-a-machine-learning-deep-learning-pilot-program-to-maximize-the-use-of-its-digital-collection-library-is-looking-for-r/	https://www.infodocket.com/2019/06/13/library-of-congress-posts-solicitation-for-a-machine-learning-deep-learning-pilot-program-to-maximize-the-use-of-its-digital-collection-library-is-looking-for-r/	NN
07-kim-ai	377	33	https://www.blog.google/products/assistant/interpreter-mode-brings-real-time-translation-your-phone/	https://www.blog.google/products/assistant/interpreter-mode-brings-real-time-translation-your-phone/	NNP
07-kim-ai	377	34	https://www.blog.google/products/assistant/interpreter-mode-brings-real-time-translation-your-phone/	https://www.blog.google/products/assistant/interpreter-mode-brings-real-time-translation-your-phone/	NNP
07-kim-ai	377	35	Kim	Kim	NNP
07-kim-ai	377	36	85	85	CD
07-kim-ai	377	37	MifBMi2`T`2i2`@KQ/2@#`BM;b@`2	MifBMi2`T`2i2`@KQ/2@#`BM;b@`2	NNP
07-kim-ai	377	38	�	�	NNP
07-kim-ai	377	39	H@iBK2@i`	H@iBK2@i`	NNP
07-kim-ai	377	40	�	�	NNP
07-kim-ai	377	41	MbH	MbH	NNP
07-kim-ai	377	42	�	�	NNP
07-kim-ai	377	43	iBQM@vQm`@T?QM2f	ibqm@vqm`@t?qm2f	NN
07-kim-ai	377	44	.	.	.
07-kim-ai	378	1	Sharkey	Sharkey	NNP
07-kim-ai	378	2	,	,	,
07-kim-ai	378	3	Noel	Noel	NNP
07-kim-ai	378	4	.	.	.
07-kim-ai	379	1	2012	2012	CD
07-kim-ai	379	2	.	.	.
07-kim-ai	380	1	“	"	``
07-kim-ai	380	2	Killing	kill	VBG
07-kim-ai	380	3	Made	make	VBN
07-kim-ai	380	4	Easy	Easy	NNP
07-kim-ai	380	5	:	:	:
07-kim-ai	380	6	From	from	IN
07-kim-ai	380	7	Joysticks	Joysticks	NNP
07-kim-ai	380	8	to	to	IN
07-kim-ai	380	9	Politics	politic	NNS
07-kim-ai	380	10	.	.	.
07-kim-ai	380	11	”	"	''
07-kim-ai	380	12	In	in	IN
07-kim-ai	380	13	Robot	Robot	NNP
07-kim-ai	380	14	Ethics	Ethics	NNP
07-kim-ai	380	15	:	:	:
07-kim-ai	380	16	The	the	DT
07-kim-ai	380	17	Ethical	Ethical	NNP
07-kim-ai	380	18	and	and	CC
07-kim-ai	380	19	Social	Social	NNP
07-kim-ai	380	20	Implications	Implications	NNPS
07-kim-ai	380	21	of	of	IN
07-kim-ai	380	22	Robotics	Robotics	NNPS
07-kim-ai	380	23	,	,	,
07-kim-ai	380	24	edited	edit	VBN
07-kim-ai	380	25	by	by	IN
07-kim-ai	380	26	Patrick	Patrick	NNP
07-kim-ai	380	27	Lin	Lin	NNP
07-kim-ai	380	28	,	,	,
07-kim-ai	380	29	Keith	Keith	NNP
07-kim-ai	380	30	Abney	Abney	NNP
07-kim-ai	380	31	,	,	,
07-kim-ai	380	32	and	and	CC
07-kim-ai	380	33	George	George	NNP
07-kim-ai	380	34	A.	A.	NNP
07-kim-ai	380	35	Bekey	Bekey	NNP
07-kim-ai	380	36	,	,	,
07-kim-ai	380	37	111–28	111–28	CD
07-kim-ai	380	38	.	.	.
07-kim-ai	381	1	Intelligent	intelligent	JJ
07-kim-ai	381	2	Robotics	Robotics	NNPS
07-kim-ai	381	3	and	and	CC
07-kim-ai	381	4	Autonomous	Autonomous	NNP
07-kim-ai	381	5	Agents	Agents	NNPS
07-kim-ai	381	6	.	.	.
07-kim-ai	382	1	Cambridge	Cambridge	NNP
07-kim-ai	382	2	,	,	,
07-kim-ai	382	3	Mass.	Massachusetts	NNP
07-kim-ai	383	1	:	:	:
07-kim-ai	383	2	MIT	MIT	NNP
07-kim-ai	383	3	Press	Press	NNP
07-kim-ai	383	4	.	.	.
07-kim-ai	384	1	Singer	singer	NN
07-kim-ai	384	2	,	,	,
07-kim-ai	384	3	Peter	Peter	NNP
07-kim-ai	384	4	.	.	.
07-kim-ai	385	1	2005	2005	CD
07-kim-ai	385	2	.	.	.
07-kim-ai	386	1	“	"	``
07-kim-ai	386	2	Ethics	Ethics	NNPS
07-kim-ai	386	3	and	and	CC
07-kim-ai	386	4	Intuitions	Intuitions	NNPS
07-kim-ai	386	5	.	.	.
07-kim-ai	386	6	”	"	''
07-kim-ai	386	7	The	the	DT
07-kim-ai	386	8	Journal	Journal	NNP
07-kim-ai	386	9	of	of	IN
07-kim-ai	386	10	Ethics	Ethics	NNP
07-kim-ai	386	11	9	9	CD
07-kim-ai	386	12	(	(	-LRB-
07-kim-ai	386	13	3/4	3/4	CD
07-kim-ai	386	14	)	)	-RRB-
07-kim-ai	386	15	:	:	:
07-kim-ai	386	16	331–52	331–52	CD
07-kim-ai	386	17	.	.	.
07-kim-ai	387	1	Sinnott	Sinnott	NNP
07-kim-ai	387	2	-	-	HYPH
07-kim-ai	387	3	Armstrong	Armstrong	NNP
07-kim-ai	387	4	,	,	,
07-kim-ai	387	5	Walter	Walter	NNP
07-kim-ai	387	6	.	.	.
07-kim-ai	388	1	2019	2019	CD
07-kim-ai	388	2	.	.	.
07-kim-ai	389	1	“	"	``
07-kim-ai	389	2	Consequentialism	consequentialism	NN
07-kim-ai	389	3	.	.	.
07-kim-ai	389	4	”	"	''
07-kim-ai	389	5	In	in	IN
07-kim-ai	389	6	The	the	DT
07-kim-ai	389	7	Stanford	Stanford	NNP
07-kim-ai	389	8	Encyclopedia	Encyclopedia	NNP
07-kim-ai	389	9	of	of	IN
07-kim-ai	389	10	Phi-	Phi-	NNP
07-kim-ai	389	11	losophy	losophy	NN
07-kim-ai	389	12	,	,	,
07-kim-ai	389	13	edited	edit	VBN
07-kim-ai	389	14	by	by	IN
07-kim-ai	389	15	Edward	Edward	NNP
07-kim-ai	389	16	N.	N.	NNP
07-kim-ai	389	17	Zalta	Zalta	NNP
07-kim-ai	389	18	,	,	,
07-kim-ai	389	19	Summer	Summer	NNP
07-kim-ai	389	20	2019	2019	CD
07-kim-ai	389	21	.	.	.
07-kim-ai	390	1	Metaphysics	Metaphysics	NNP
07-kim-ai	390	2	Research	Research	NNP
07-kim-ai	390	3	Lab	Lab	NNP
07-kim-ai	390	4	,	,	,
07-kim-ai	390	5	Stanford	Stanford	NNP
07-kim-ai	390	6	University	University	NNP
07-kim-ai	390	7	.	.	.
07-kim-ai	391	1	?	?	.
07-kim-ai	391	2	iiTb	iitb	UH
07-kim-ai	391	3	,	,	,
07-kim-ai	391	4	ffTH	ffTH	NNP
07-kim-ai	391	5	�	�	NNP
07-kim-ai	391	6	iQXbi	iQXbi	NNS
07-kim-ai	391	7	�	�	NNP
07-kim-ai	391	8	M7Q`/X2	m7q`/x2	NN
07-kim-ai	391	9	/	/	SYM
07-kim-ai	391	10	mf	mf	IN
07-kim-ai	391	11	�	�	NNP
07-kim-ai	391	12	`+?Bp2bfbmKkyRNf2Mi`B2bf+QMb	`+?Bp2bfbmKkyRNf2Mi`B2bf+QMb	NNP
07-kim-ai	391	13	2[m2MiB	2[m2mib	CD
07-kim-ai	391	14	�	�	VBZ
07-kim-ai	391	15	HBbKf	hbbkf	PRP
07-kim-ai	391	16	.	.	.
07-kim-ai	392	1	Stanley	Stanley	NNP
07-kim-ai	392	2	,	,	,
07-kim-ai	392	3	Jay	Jay	NNP
07-kim-ai	392	4	.	.	.
07-kim-ai	393	1	2017	2017	CD
07-kim-ai	393	2	.	.	.
07-kim-ai	394	1	“	"	``
07-kim-ai	394	2	Pitfalls	pitfall	NNS
07-kim-ai	394	3	of	of	IN
07-kim-ai	394	4	Artificial	Artificial	NNP
07-kim-ai	394	5	Intelligence	Intelligence	NNP
07-kim-ai	394	6	Decisionmaking	Decisionmaking	NNP
07-kim-ai	394	7	Highlighted	highlight	VBN
07-kim-ai	394	8	In	in	IN
07-kim-ai	394	9	Idaho	Idaho	NNP
07-kim-ai	394	10	ACLU	ACLU	NNP
07-kim-ai	394	11	Case	Case	NNP
07-kim-ai	394	12	.	.	.
07-kim-ai	394	13	”	"	''
07-kim-ai	394	14	American	American	NNP
07-kim-ai	394	15	Civil	Civil	NNP
07-kim-ai	394	16	Liberties	Liberties	NNPS
07-kim-ai	394	17	Union	Union	NNP
07-kim-ai	394	18	(	(	-LRB-
07-kim-ai	394	19	blog	blog	NN
07-kim-ai	394	20	)	)	-RRB-
07-kim-ai	394	21	.	.	.
07-kim-ai	395	1	June	June	NNP
07-kim-ai	395	2	2	2	CD
07-kim-ai	395	3	,	,	,
07-kim-ai	395	4	2017	2017	CD
07-kim-ai	395	5	.	.	.
07-kim-ai	396	1	?	?	.
07-kim-ai	396	2	iiTb	iitb	UH
07-kim-ai	396	3	,	,	,
07-kim-ai	396	4	ffrrrX	ffrrrX	NNP
07-kim-ai	396	5	�	�	NNP
07-kim-ai	396	6	+HmXQ`;f	+hmxq`;f	NN
07-kim-ai	396	7	#	#	NN
07-kim-ai	396	8	HQ;fT`Bp	hq;ft`bp	CD
07-kim-ai	396	9	�	�	NNS
07-kim-ai	396	10	+v@i2+?MQHQ;vfTBi7	+v@i2+?MQHQ;vfTBi7	VBZ
07-kim-ai	396	11	�	�	JJ
07-kim-ai	396	12	HHb@	HHb@	NNS
07-kim-ai	396	13	�	�	.
07-kim-ai	396	14	`iB7B+B	`iB7B+B	NNP
07-kim-ai	396	15	�	�	NNP
07-kim-ai	396	16	H@BMi2HHB;2M+2@/2+BbBQM	H@BMi2HHB;2M+2@/2+BbBQM	NNP
07-kim-ai	396	17	K	K	NNP
07-kim-ai	396	18	�	�	NNP
07-kim-ai	396	19	FBM;@?B;?HB;?i2/@B/	FBM;@?B;?HB;?i2/@B/	NNP
07-kim-ai	396	20	�	�	NNP
07-kim-ai	396	21	?Q@	?Q@	.
07-kim-ai	396	22	�	�	NNS
07-kim-ai	396	23	+Hm@+	+Hm@+	NNS
07-kim-ai	396	24	�	�	NNP
07-kim-ai	396	25	b2	b2	NN
07-kim-ai	396	26	.	.	.
07-kim-ai	397	1	Talley	Talley	NNP
07-kim-ai	397	2	,	,	,
07-kim-ai	397	3	Nancy	Nancy	NNP
07-kim-ai	397	4	B.	B.	NNP
07-kim-ai	398	1	2016	2016	CD
07-kim-ai	398	2	.	.	.
07-kim-ai	399	1	“	"	``
07-kim-ai	399	2	Imagining	imagine	VBG
07-kim-ai	399	3	the	the	DT
07-kim-ai	399	4	Use	Use	NNP
07-kim-ai	399	5	of	of	IN
07-kim-ai	399	6	Intelligent	Intelligent	NNP
07-kim-ai	399	7	Agents	Agents	NNPS
07-kim-ai	399	8	and	and	CC
07-kim-ai	399	9	Artificial	Artificial	NNP
07-kim-ai	399	10	Intelligence	Intelligence	NNP
07-kim-ai	399	11	in	in	IN
07-kim-ai	399	12	Academic	Academic	NNP
07-kim-ai	399	13	Law	Law	NNP
07-kim-ai	399	14	Libraries	Libraries	NNPS
07-kim-ai	399	15	.	.	.
07-kim-ai	399	16	”	"	''
07-kim-ai	399	17	Law	Law	NNP
07-kim-ai	399	18	Library	Library	NNP
07-kim-ai	399	19	Journal	Journal	NNP
07-kim-ai	399	20	108	108	CD
07-kim-ai	399	21	(	(	-LRB-
07-kim-ai	399	22	3	3	CD
07-kim-ai	399	23	)	)	-RRB-
07-kim-ai	399	24	:	:	:
07-kim-ai	399	25	383–402	383–402	CD
07-kim-ai	399	26	.	.	.
07-kim-ai	400	1	Tashea	Tashea	NNP
07-kim-ai	400	2	,	,	,
07-kim-ai	400	3	Jason	Jason	NNP
07-kim-ai	400	4	.	.	.
07-kim-ai	401	1	2017	2017	CD
07-kim-ai	401	2	.	.	.
07-kim-ai	402	1	“	"	``
07-kim-ai	402	2	Courts	court	NNS
07-kim-ai	402	3	Are	be	VBP
07-kim-ai	402	4	Using	use	VBG
07-kim-ai	402	5	AI	ai	NN
07-kim-ai	402	6	to	to	IN
07-kim-ai	402	7	Sentence	Sentence	NNP
07-kim-ai	402	8	Criminals	Criminals	NNPS
07-kim-ai	402	9	.	.	.
07-kim-ai	403	1	That	that	DT
07-kim-ai	403	2	Must	Must	MD
07-kim-ai	403	3	Stop	stop	VB
07-kim-ai	403	4	Now	now	RB
07-kim-ai	403	5	.	.	.
07-kim-ai	403	6	”	"	''
07-kim-ai	403	7	Wired	wire	VBN
07-kim-ai	403	8	,	,	,
07-kim-ai	403	9	April	April	NNP
07-kim-ai	403	10	17	17	CD
07-kim-ai	403	11	,	,	,
07-kim-ai	403	12	2017	2017	CD
07-kim-ai	403	13	.	.	.
07-kim-ai	404	1	?	?	.
07-kim-ai	404	2	iiTb	iitb	UH
07-kim-ai	404	3	,	,	,
07-kim-ai	404	4	ffrrrXrB`2	ffrrrXrB`2	NNP
07-kim-ai	404	5	/	/	SYM
07-kim-ai	404	6	X+QKfkyRdfy9f+Qm`ib@mbBM;@	X+QKfkyRdfy9f+Qm`ib@mbBM;@	NNP
07-kim-ai	404	7	�	�	NNP
07-kim-ai	404	8	B@b2	B@b2	NNP
07-kim-ai	404	9	Mi2M+2@+`BKBM	Mi2M+2@+`BKBM	NNP
07-kim-ai	404	10	�	�	NNP
07-kim-ai	404	11	Hb@Kmbi@biQT@MQrf	hb@kmbi@biqt@mqrf	RB
07-kim-ai	404	12	.	.	.
07-kim-ai	405	1	Tegmark	Tegmark	NNP
07-kim-ai	405	2	,	,	,
07-kim-ai	405	3	Max	Max	NNP
07-kim-ai	405	4	.	.	.
07-kim-ai	406	1	2017	2017	CD
07-kim-ai	406	2	.	.	.
07-kim-ai	407	1	Life	life	NN
07-kim-ai	407	2	3.0	3.0	CD
07-kim-ai	407	3	:	:	:
07-kim-ai	407	4	Being	be	VBG
07-kim-ai	407	5	Human	human	NN
07-kim-ai	407	6	in	in	IN
07-kim-ai	407	7	the	the	DT
07-kim-ai	407	8	Age	Age	NNP
07-kim-ai	407	9	of	of	IN
07-kim-ai	407	10	Artificial	Artificial	NNP
07-kim-ai	407	11	Intelligence	Intelligence	NNP
07-kim-ai	407	12	.	.	.
07-kim-ai	408	1	New	New	NNP
07-kim-ai	408	2	York	York	NNP
07-kim-ai	408	3	:	:	:
07-kim-ai	408	4	Alfred	Alfred	NNP
07-kim-ai	408	5	Knopf	Knopf	NNP
07-kim-ai	408	6	.	.	.
07-kim-ai	409	1	Thomson	Thomson	NNP
07-kim-ai	409	2	,	,	,
07-kim-ai	409	3	Judith	Judith	NNP
07-kim-ai	409	4	Jarvis	Jarvis	NNP
07-kim-ai	409	5	.	.	.
07-kim-ai	410	1	1976	1976	CD
07-kim-ai	410	2	.	.	.
07-kim-ai	411	1	“	"	``
07-kim-ai	411	2	Killing	kill	VBG
07-kim-ai	411	3	,	,	,
07-kim-ai	411	4	Letting	Letting	NNP
07-kim-ai	411	5	Die	Die	NNP
07-kim-ai	411	6	,	,	,
07-kim-ai	411	7	and	and	CC
07-kim-ai	411	8	the	the	DT
07-kim-ai	411	9	Trolley	Trolley	NNP
07-kim-ai	411	10	Problem	Problem	NNP
07-kim-ai	411	11	.	.	.
07-kim-ai	411	12	”	"	''
07-kim-ai	411	13	The	the	DT
07-kim-ai	411	14	Monist	Monist	NNP
07-kim-ai	411	15	59	59	CD
07-kim-ai	411	16	(	(	-LRB-
07-kim-ai	411	17	2	2	CD
07-kim-ai	411	18	)	)	-RRB-
07-kim-ai	411	19	:	:	:
07-kim-ai	411	20	204–17	204–17	CD
07-kim-ai	411	21	.	.	.
07-kim-ai	412	1	Turek	Turek	NNP
07-kim-ai	412	2	,	,	,
07-kim-ai	412	3	Matt	Matt	NNP
07-kim-ai	412	4	.	.	.
07-kim-ai	413	1	n.d	n.d	NNP
07-kim-ai	413	2	.	.	.
07-kim-ai	413	3	“	"	``
07-kim-ai	413	4	Explainable	Explainable	NNP
07-kim-ai	413	5	Artificial	Artificial	NNP
07-kim-ai	413	6	Intelligence	Intelligence	NNP
07-kim-ai	413	7	.	.	.
07-kim-ai	413	8	”	"	''
07-kim-ai	413	9	Defense	Defense	NNP
07-kim-ai	413	10	Advanced	Advanced	NNP
07-kim-ai	413	11	Research	Research	NNP
07-kim-ai	413	12	Projects	Projects	NNPS
07-kim-ai	413	13	Agency	Agency	NNP
07-kim-ai	413	14	.	.	.
07-kim-ai	414	1	?	?	.
07-kim-ai	414	2	iiTb	iitb	UH
07-kim-ai	414	3	,	,	,
07-kim-ai	414	4	ffrrrX/	ffrrrX/	NNP
07-kim-ai	414	5	�	�	NNP
07-kim-ai	414	6	`T	`t	CD
07-kim-ai	414	7	�	�	JJ
07-kim-ai	414	8	XKBHfT`Q;`	xkbhft`q;`	NN
07-kim-ai	414	9	�	�	NN
07-kim-ai	414	10	Kf2tTH	Kf2tTH	,
07-kim-ai	414	11	�	�	,
07-kim-ai	414	12	BM	bm	NN
07-kim-ai	414	13	�	�	NN
07-kim-ai	414	14	#H2@	#H2@	NNP
07-kim-ai	414	15	�	�	.
07-kim-ai	414	16	`iB7B+B	`ib7b+b	JJ
07-kim-ai	414	17	�	�	JJ
07-kim-ai	414	18	H@BMi2H	H@BMi2H	NNP
07-kim-ai	414	19	HB;2M+2	hb;2m+2	NN
07-kim-ai	414	20	.	.	.
07-kim-ai	415	1	Vallor	Vallor	NNP
07-kim-ai	415	2	,	,	,
07-kim-ai	415	3	Shannon	Shannon	NNP
07-kim-ai	415	4	.	.	.
07-kim-ai	416	1	2015	2015	CD
07-kim-ai	416	2	.	.	.
07-kim-ai	417	1	“	"	``
07-kim-ai	417	2	Moral	Moral	NNP
07-kim-ai	417	3	Deskilling	Deskilling	NNP
07-kim-ai	417	4	and	and	CC
07-kim-ai	417	5	Upskilling	upskille	VBG
07-kim-ai	417	6	in	in	IN
07-kim-ai	417	7	a	a	DT
07-kim-ai	417	8	New	New	NNP
07-kim-ai	417	9	Machine	Machine	NNP
07-kim-ai	417	10	Age	Age	NNP
07-kim-ai	417	11	:	:	:
07-kim-ai	417	12	Reflections	reflection	NNS
07-kim-ai	417	13	on	on	IN
07-kim-ai	417	14	the	the	DT
07-kim-ai	417	15	Ambiguous	ambiguous	JJ
07-kim-ai	417	16	Future	Future	NNP
07-kim-ai	417	17	of	of	IN
07-kim-ai	417	18	Character	Character	NNP
07-kim-ai	417	19	.	.	.
07-kim-ai	417	20	”	"	''
07-kim-ai	417	21	Philosophy	Philosophy	NNP
07-kim-ai	417	22	&	&	CC
07-kim-ai	417	23	Technology	Technology	NNP
07-kim-ai	417	24	28	28	CD
07-kim-ai	417	25	(	(	-LRB-
07-kim-ai	417	26	1	1	CD
07-kim-ai	417	27	)	)	-RRB-
07-kim-ai	417	28	:	:	:
07-kim-ai	417	29	107–24	107–24	CD
07-kim-ai	417	30	.	.	.
07-kim-ai	418	1	?	?	.
07-kim-ai	418	2	iiTb	iitb	UH
07-kim-ai	418	3	,	,	,
07-kim-ai	418	4	ff	ff	FW
07-kim-ai	418	5	/	/	SYM
07-kim-ai	418	6	QBXQ`;fRyXRyydfbRjj9d@yR9@yR8e@N.	QBXQ`;fRyXRyydfbRjj9d@yR9@yR8e@N.	NNP
07-kim-ai	418	7	Wallach	Wallach	NNP
07-kim-ai	418	8	,	,	,
07-kim-ai	418	9	Wendell	Wendell	NNP
07-kim-ai	418	10	.	.	.
07-kim-ai	419	1	2009	2009	CD
07-kim-ai	419	2	.	.	.
07-kim-ai	420	1	Moral	moral	JJ
07-kim-ai	420	2	Machines	Machines	NNPS
07-kim-ai	420	3	:	:	:
07-kim-ai	420	4	Teaching	teaching	NN
07-kim-ai	420	5	Robots	robot	NNS
07-kim-ai	420	6	Right	right	RB
07-kim-ai	420	7	from	from	IN
07-kim-ai	420	8	Wrong	wrong	NN
07-kim-ai	420	9	.	.	.
07-kim-ai	421	1	Oxford	oxford	NN
07-kim-ai	421	2	:	:	:
07-kim-ai	421	3	Oxford	Oxford	NNP
07-kim-ai	421	4	University	University	NNP
07-kim-ai	421	5	Press	Press	NNP
07-kim-ai	421	6	.	.	.
07-kim-ai	422	1	Welch	Welch	NNP
07-kim-ai	422	2	,	,	,
07-kim-ai	422	3	Chris	Chris	NNP
07-kim-ai	422	4	.	.	.
07-kim-ai	423	1	2018	2018	CD
07-kim-ai	423	2	.	.	.
07-kim-ai	424	1	“	"	``
07-kim-ai	424	2	Google	Google	NNP
07-kim-ai	424	3	Just	just	RB
07-kim-ai	424	4	Gave	give	VBD
07-kim-ai	424	5	a	a	DT
07-kim-ai	424	6	Stunning	Stunning	NNP
07-kim-ai	424	7	Demo	Demo	NNP
07-kim-ai	424	8	of	of	IN
07-kim-ai	424	9	Assistant	Assistant	NNP
07-kim-ai	424	10	Making	make	VBG
07-kim-ai	424	11	an	an	DT
07-kim-ai	424	12	Actual	actual	JJ
07-kim-ai	424	13	Phone	Phone	NNP
07-kim-ai	424	14	Call	call	NN
07-kim-ai	424	15	.	.	.
07-kim-ai	424	16	”	"	''
07-kim-ai	424	17	The	the	DT
07-kim-ai	424	18	Verge	Verge	NNP
07-kim-ai	424	19	.	.	.
07-kim-ai	425	1	May	May	NNP
07-kim-ai	425	2	8	8	CD
07-kim-ai	425	3	,	,	,
07-kim-ai	425	4	2018	2018	CD
07-kim-ai	425	5	.	.	.
07-kim-ai	426	1	?	?	.
07-kim-ai	426	2	iiTb	iitb	UH
07-kim-ai	426	3	,	,	,
07-kim-ai	426	4	ffrrrXi?2p2`;2X+QKfkyR3f8f3fRdjjkydy	ffrrrXi?2p2`;2X+QKfkyR3f8f3fRdjjkydy	NFP
07-kim-ai	426	5	f;QQ;H2@	f;qq;h2@	ADD
07-kim-ai	426	6	�	�	NNP
07-kim-ai	426	7	bbBbi	bbBbi	NNP
07-kim-ai	426	8	�	�	NNP
07-kim-ai	426	9	Mi@K	mi@k	NN
07-kim-ai	426	10	�	�	NNP
07-kim-ai	426	11	F2b@T?QM2@+	F2b@T?QM2@+	NNP
07-kim-ai	426	12	�	�	NNP
07-kim-ai	426	13	HH@/2KQ@/mTH2t@BQ@kyR3	HH@/2KQ@/mTH2t@BQ@kyR3	NNP
07-kim-ai	426	14	.	.	.
07-kim-ai	427	1	https://www.blog.google/products/assistant/interpreter-mode-brings-real-time-translation-your-phone/	https://www.blog.google/products/assistant/interpreter-mode-brings-real-time-translation-your-phone/	NNP
07-kim-ai	427	2	https://www.blog.google/products/assistant/interpreter-mode-brings-real-time-translation-your-phone/	https://www.blog.google/products/assistant/interpreter-mode-brings-real-time-translation-your-phone/	NNP
07-kim-ai	427	3	https://plato.stanford.edu/archives/sum2019/entries/consequentialism/	https://plato.stanford.edu/archives/sum2019/entries/consequentialism/	JJ
07-kim-ai	427	4	https://plato.stanford.edu/archives/sum2019/entries/consequentialism/	https://plato.stanford.edu/archives/sum2019/entries/consequentialism/	NNS
07-kim-ai	427	5	https://www.aclu.org/blog/privacy-technology/pitfalls-artificial-intelligence-decisionmaking-highlighted-idaho-aclu-case	https://www.aclu.org/blog/privacy-technology/pitfalls-artificial-intelligence-decisionmaking-highlighted-idaho-aclu-case	NNS
07-kim-ai	427	6	https://www.aclu.org/blog/privacy-technology/pitfalls-artificial-intelligence-decisionmaking-highlighted-idaho-aclu-case	https://www.aclu.org/blog/privacy-technology/pitfalls-artificial-intelligence-decisionmaking-highlighted-idaho-aclu-case	NNS
07-kim-ai	427	7	https://www.aclu.org/blog/privacy-technology/pitfalls-artificial-intelligence-decisionmaking-highlighted-idaho-aclu-case	https://www.aclu.org/blog/privacy-technology/pitfalls-artificial-intelligence-decisionmaking-highlighted-idaho-aclu-case	NNP
07-kim-ai	427	8	https://www.wired.com/2017/04/courts-using-ai-sentence-criminals-must-stop-now/	https://www.wired.com/2017/04/courts-using-ai-sentence-criminals-must-stop-now/	CD
07-kim-ai	427	9	https://www.wired.com/2017/04/courts-using-ai-sentence-criminals-must-stop-now/	https://www.wired.com/2017/04/courts-using-ai-sentence-criminals-must-stop-now/	CD
07-kim-ai	427	10	https://www.darpa.mil/program/explainable-artificial-intelligence	https://www.darpa.mil/program/explainable-artificial-intelligence	NN
07-kim-ai	427	11	https://www.darpa.mil/program/explainable-artificial-intelligence	https://www.darpa.mil/program/explainable-artificial-intelligence	NN
07-kim-ai	427	12	https://doi.org/10.1007/s13347-014-0156-9	https://doi.org/10.1007/s13347-014-0156-9	CD
07-kim-ai	427	13	https://doi.org/10.1007/s13347-014-0156-9	https://doi.org/10.1007/s13347-014-0156-9	CD
07-kim-ai	427	14	https://www.theverge.com/2018/5/8/17332070/google-assistant-makes-phone-call-demo-duplex-io-2018	https://www.theverge.com/2018/5/8/17332070/google-assistant-makes-phone-call-demo-duplex-io-2018	NNP
07-kim-ai	427	15	https://www.theverge.com/2018/5/8/17332070/google-assistant-makes-phone-call-demo-duplex-io-2018	https://www.theverge.com/2018/5/8/17332070/google-assistant-makes-phone-call-demo-duplex-io-2018	NNP