id	sid	tid	token	lemma	pos
kim	1	1	Bohyun	Bohyun	NNP
kim	1	2	Kim	Kim	NNP
kim	1	3	,	,	,
kim	1	4	AI	AI	NNP
kim	1	5	and	and	CC
kim	1	6	Its	-PRON-	PRP$
kim	1	7	Moral	Moral	NNP
kim	1	8	Concerns	Concerns	NNPS
kim	1	9	AI	AI	NNP
kim	1	10	and	and	CC
kim	1	11	Its	-PRON-	PRP$
kim	1	12	Moral	Moral	NNP
kim	1	13	Concerns	Concerns	NNPS
kim	1	14	Bohyun	Bohyun	NNP
kim	1	15	Kim	Kim	NNP
kim	1	16	1	1	CD
kim	1	17	.	.	.
kim	2	1	Automating	automate	VBG
kim	2	2	Decisions	Decisions	NNPS
kim	2	3	and	and	CC
kim	2	4	Actions	Actions	NNPS
kim	2	5	The	the	DT
kim	2	6	goal	goal	NN
kim	2	7	of	of	IN
kim	2	8	artificial	artificial	JJ
kim	2	9	intelligence	intelligence	NN
kim	2	10	(	(	-LRB-
kim	2	11	AI	AI	NNP
kim	2	12	)	)	-RRB-
kim	2	13	as	as	IN
kim	2	14	a	a	DT
kim	2	15	discipline	discipline	NN
kim	2	16	is	be	VBZ
kim	2	17	to	to	TO
kim	2	18	create	create	VB
kim	2	19	an	an	DT
kim	2	20	artificial	artificial	JJ
kim	2	21	system	system	NN
kim	2	22	–	–	:
kim	2	23	whether	whether	IN
kim	2	24	it	-PRON-	PRP
kim	2	25	be	be	VB
kim	2	26	a	a	DT
kim	2	27	piece	piece	NN
kim	2	28	of	of	IN
kim	2	29	software	software	NN
kim	2	30	or	or	CC
kim	2	31	a	a	DT
kim	2	32	machine	machine	NN
kim	2	33	with	with	IN
kim	2	34	a	a	DT
kim	2	35	physical	physical	JJ
kim	2	36	body	body	NN
kim	2	37	–	–	:
kim	2	38	that	that	DT
kim	2	39	is	be	VBZ
kim	2	40	as	as	RB
kim	2	41	intelligent	intelligent	JJ
kim	2	42	as	as	IN
kim	2	43	a	a	DT
kim	2	44	human	human	NN
kim	2	45	in	in	IN
kim	2	46	its	-PRON-	PRP$
kim	2	47	performance	performance	NN
kim	2	48	,	,	,
kim	2	49	either	either	CC
kim	2	50	broadly	broadly	RB
kim	2	51	in	in	IN
kim	2	52	all	all	DT
kim	2	53	areas	area	NNS
kim	2	54	of	of	IN
kim	2	55	human	human	JJ
kim	2	56	activities	activity	NNS
kim	2	57	or	or	CC
kim	2	58	narrowly	narrowly	RB
kim	2	59	in	in	IN
kim	2	60	a	a	DT
kim	2	61	specific	specific	JJ
kim	2	62	activity	activity	NN
kim	2	63	,	,	,
kim	2	64	such	such	JJ
kim	2	65	as	as	IN
kim	2	66	playing	play	VBG
kim	2	67	chess	chess	NN
kim	2	68	or	or	CC
kim	2	69	driving	driving	NN
kim	2	70	.	.	.
kim	3	1	[	[	-LRB-
kim	3	2	footnoteRef:0	footnoteref:0	NN
kim	3	3	]	]	-RRB-
kim	3	4	The	the	DT
kim	3	5	actual	actual	JJ
kim	3	6	capability	capability	NN
kim	3	7	of	of	IN
kim	3	8	most	most	JJS
kim	3	9	AI	AI	NNP
kim	3	10	systems	system	NNS
kim	3	11	remained	remain	VBD
kim	3	12	far	far	RB
kim	3	13	below	below	IN
kim	3	14	this	this	DT
kim	3	15	ambitious	ambitious	JJ
kim	3	16	goal	goal	NN
kim	3	17	for	for	IN
kim	3	18	a	a	DT
kim	3	19	long	long	JJ
kim	3	20	time	time	NN
kim	3	21	.	.	.
kim	4	1	But	but	CC
kim	4	2	with	with	IN
kim	4	3	recent	recent	JJ
kim	4	4	successes	success	NNS
kim	4	5	with	with	IN
kim	4	6	machine	machine	NN
kim	4	7	learning	learning	NN
kim	4	8	and	and	CC
kim	4	9	deep	deep	JJ
kim	4	10	learning	learning	NN
kim	4	11	,	,	,
kim	4	12	the	the	DT
kim	4	13	performance	performance	NN
kim	4	14	of	of	IN
kim	4	15	some	some	DT
kim	4	16	AI	AI	NNP
kim	4	17	programs	program	NNS
kim	4	18	has	have	VBZ
kim	4	19	started	start	VBN
kim	4	20	surpassing	surpass	VBG
kim	4	21	that	that	DT
kim	4	22	of	of	IN
kim	4	23	humans	human	NNS
kim	4	24	.	.	.
kim	5	1	In	in	IN
kim	5	2	2016	2016	CD
kim	5	3	,	,	,
kim	5	4	an	an	DT
kim	5	5	AI	AI	NNP
kim	5	6	program	program	NN
kim	5	7	developed	develop	VBN
kim	5	8	with	with	IN
kim	5	9	the	the	DT
kim	5	10	deep	deep	JJ
kim	5	11	learning	learning	NN
kim	5	12	method	method	NN
kim	5	13	,	,	,
kim	5	14	AlphaGo	AlphaGo	NNP
kim	5	15	,	,	,
kim	5	16	astonished	astonish	VBD
kim	5	17	even	even	RB
kim	5	18	its	-PRON-	PRP$
kim	5	19	creators	creator	NNS
kim	5	20	by	by	IN
kim	5	21	winning	win	VBG
kim	5	22	four	four	CD
kim	5	23	out	out	IN
kim	5	24	of	of	IN
kim	5	25	five	five	CD
kim	5	26	Go	Go	NNP
kim	5	27	matches	match	NNS
kim	5	28	with	with	IN
kim	5	29	the	the	DT
kim	5	30	eighteen	eighteen	CD
kim	5	31	-	-	HYPH
kim	5	32	time	time	NN
kim	5	33	world	world	NN
kim	5	34	champion	champion	NN
kim	5	35	,	,	,
kim	5	36	 	  	_SP
kim	5	37	Sedol	Sedol	NNP
kim	5	38	Lee	Lee	NNP
kim	5	39	.	.	.
kim	6	1	[	[	-LRB-
kim	6	2	footnoteRef:1	footnoteRef:1	NNP
kim	6	3	]	]	-RRB-
kim	6	4	In	in	IN
kim	6	5	2020	2020	CD
kim	6	6	,	,	,
kim	6	7	Google	Google	NNP
kim	6	8	’s	’s	NNP
kim	6	9	DeepMind	DeepMind	NNP
kim	6	10	unveiled	unveil	VBD
kim	6	11	Atari57	Atari57	NNP
kim	6	12	,	,	,
kim	6	13	a	a	DT
kim	6	14	deep	deep	JJ
kim	6	15	reinforcement	reinforcement	NN
kim	6	16	learning	learn	VBG
kim	6	17	algorithm	algorithm	NN
kim	6	18	that	that	WDT
kim	6	19	reached	reach	VBD
kim	6	20	superhuman	superhuman	JJ
kim	6	21	levels	level	NNS
kim	6	22	of	of	IN
kim	6	23	play	play	NN
kim	6	24	in	in	IN
kim	6	25	57	57	CD
kim	6	26	classic	classic	JJ
kim	6	27	Atari	Atari	NNP
kim	6	28	games	game	NNS
kim	6	29	.	.	.
kim	7	1	[	[	-LRB-
kim	7	2	footnoteRef:2	footnoteRef:2	VBN
kim	7	3	]	]	-RRB-
kim	7	4	[	[	-LRB-
kim	7	5	0	0	CD
kim	7	6	:	:	:
kim	7	7	Note	note	VB
kim	7	8	that	that	IN
kim	7	9	by	by	IN
kim	7	10	‘	'	``
kim	7	11	as	as	RB
kim	7	12	intelligent	intelligent	JJ
kim	7	13	as	as	IN
kim	7	14	a	a	DT
kim	7	15	human	human	NN
kim	7	16	,	,	,
kim	7	17	’	'	''
kim	7	18	I	-PRON-	PRP
kim	7	19	simply	simply	RB
kim	7	20	mean	mean	VBP
kim	7	21	human	human	JJ
kim	7	22	-	-	HYPH
kim	7	23	level	level	NN
kim	7	24	AI	ai	NN
kim	7	25	,	,	,
kim	7	26	or	or	CC
kim	7	27	more	more	RBR
kim	7	28	accurately	accurately	RB
kim	7	29	,	,	,
kim	7	30	AI	AI	NNP
kim	7	31	at	at	IN
kim	7	32	human	human	JJ
kim	7	33	-	-	HYPH
kim	7	34	level	level	NN
kim	7	35	performance	performance	NN
kim	7	36	.	.	.
kim	8	1	By	by	IN
kim	8	2	‘	'	``
kim	8	3	as	as	RB
kim	8	4	intelligent	intelligent	JJ
kim	8	5	as	as	IN
kim	8	6	a	a	DT
kim	8	7	human	human	NN
kim	8	8	,	,	,
kim	8	9	’	'	''
kim	8	10	I	-PRON-	PRP
kim	8	11	do	do	VBP
kim	8	12	not	not	RB
kim	8	13	mean	mean	VB
kim	8	14	general(/strong	general(/strong	.
kim	8	15	)	)	-RRB-
kim	8	16	AI	AI	NNP
kim	8	17	.	.	.
kim	9	1	General	General	NNP
kim	9	2	AI	AI	NNP
kim	9	3	–	–	:
kim	9	4	also	also	RB
kim	9	5	known	know	VBN
kim	9	6	as	as	IN
kim	9	7	‘	'	``
kim	9	8	artificial	artificial	JJ
kim	9	9	general	general	JJ
kim	9	10	intelligence	intelligence	NN
kim	9	11	(	(	-LRB-
kim	9	12	AGI	AGI	NNP
kim	9	13	)	)	-RRB-
kim	9	14	’	'	''
kim	9	15	and	and	CC
kim	9	16	‘	'	``
kim	9	17	strong	strong	JJ
kim	9	18	AI	AI	NNP
kim	9	19	’	'	''
kim	9	20	–	–	:
kim	9	21	refers	refer	VBZ
kim	9	22	to	to	IN
kim	9	23	the	the	DT
kim	9	24	ability	ability	NN
kim	9	25	to	to	TO
kim	9	26	adapt	adapt	VB
kim	9	27	to	to	TO
kim	9	28	achieve	achieve	VB
kim	9	29	any	any	DT
kim	9	30	goals	goal	NNS
kim	9	31	;	;	:
kim	9	32	it	-PRON-	PRP
kim	9	33	does	do	VBZ
kim	9	34	not	not	RB
kim	9	35	refer	refer	VB
kim	9	36	to	to	IN
kim	9	37	human	human	JJ
kim	9	38	-	-	HYPH
kim	9	39	level	level	NN
kim	9	40	performance	performance	NN
kim	9	41	in	in	IN
kim	9	42	achieving	achieve	VBG
kim	9	43	one	one	CD
kim	9	44	particular	particular	JJ
kim	9	45	goal	goal	NN
kim	9	46	.	.	.
kim	10	1	An	an	DT
kim	10	2	AI	AI	NNP
kim	10	3	system	system	NN
kim	10	4	developed	develop	VBD
kim	10	5	to	to	TO
kim	10	6	perform	perform	VB
kim	10	7	one	one	CD
kim	10	8	specific	specific	JJ
kim	10	9	type	type	NN
kim	10	10	of	of	IN
kim	10	11	activity	activity	NN
kim	10	12	or	or	CC
kim	10	13	activities	activity	NNS
kim	10	14	in	in	IN
kim	10	15	one	one	CD
kim	10	16	specific	specific	JJ
kim	10	17	domain	domain	NN
kim	10	18	is	be	VBZ
kim	10	19	called	call	VBN
kim	10	20	a	a	DT
kim	10	21	‘	'	``
kim	10	22	narrow	narrow	JJ
kim	10	23	(	(	-LRB-
kim	10	24	/weak	/weak	NFP
kim	10	25	)	)	-RRB-
kim	10	26	AI	AI	NNP
kim	10	27	’	'	''
kim	10	28	system	system	NN
kim	10	29	.	.	.
kim	10	30	]	]	-RRB-
kim	11	1	[	[	-LRB-
kim	11	2	1	1	CD
kim	11	3	:	:	:
kim	11	4	AlphaGo	alphago	PRP
kim	11	5	can	can	MD
kim	11	6	be	be	VB
kim	11	7	said	say	VBN
kim	11	8	to	to	TO
kim	11	9	be	be	VB
kim	11	10	“	"	``
kim	11	11	as	as	RB
kim	11	12	intelligent	intelligent	JJ
kim	11	13	as	as	IN
kim	11	14	humans	human	NNS
kim	11	15	,	,	,
kim	11	16	”	"	''
kim	11	17	but	but	CC
kim	11	18	only	only	RB
kim	11	19	in	in	IN
kim	11	20	playing	play	VBG
kim	11	21	Go	Go	NNP
kim	11	22	,	,	,
kim	11	23	where	where	WRB
kim	11	24	it	-PRON-	PRP
kim	11	25	exceeds	exceed	VBZ
kim	11	26	human	human	JJ
kim	11	27	capability	capability	NN
kim	11	28	.	.	.
kim	12	1	So	so	CC
kim	12	2	,	,	,
kim	12	3	it	-PRON-	PRP
kim	12	4	does	do	VBZ
kim	12	5	not	not	RB
kim	12	6	qualify	qualify	VB
kim	12	7	as	as	IN
kim	12	8	general	general	JJ
kim	12	9	/	/	SYM
kim	12	10	strong	strong	JJ
kim	12	11	AI	AI	NNP
kim	12	12	in	in	IN
kim	12	13	spite	spite	NN
kim	12	14	of	of	IN
kim	12	15	its	-PRON-	PRP$
kim	12	16	human	human	JJ
kim	12	17	-	-	HYPH
kim	12	18	level	level	NN
kim	12	19	intelligence	intelligence	NN
kim	12	20	in	in	IN
kim	12	21	Go	go	NN
kim	12	22	-	-	HYPH
kim	12	23	playing	playing	NN
kim	12	24	.	.	.
kim	13	1	It	-PRON-	PRP
kim	13	2	is	be	VBZ
kim	13	3	to	to	TO
kim	13	4	be	be	VB
kim	13	5	noted	note	VBN
kim	13	6	that	that	IN
kim	13	7	general(/strong	general(/strong	NN
kim	13	8	)	)	-RRB-
kim	13	9	AI	AI	NNP
kim	13	10	and	and	CC
kim	13	11	narrow(/weak	narrow(/weak	JJ
kim	13	12	)	)	-RRB-
kim	13	13	AI	AI	NNP
kim	13	14	signify	signify	VB
kim	13	15	the	the	DT
kim	13	16	difference	difference	NN
kim	13	17	in	in	IN
kim	13	18	the	the	DT
kim	13	19	scope	scope	NN
kim	13	20	of	of	IN
kim	13	21	AI	AI	NNP
kim	13	22	capability	capability	NN
kim	13	23	.	.	.
kim	14	1	General(/strong	General(/strong	NFP
kim	14	2	)	)	-RRB-
kim	14	3	AI	AI	NNP
kim	14	4	is	be	VBZ
kim	14	5	a	a	DT
kim	14	6	broader	broad	JJR
kim	14	7	concept	concept	NN
kim	14	8	than	than	IN
kim	14	9	human	human	NN
kim	14	10	-	-	HYPH
kim	14	11	like	like	JJ
kim	14	12	intelligence	intelligence	NN
kim	14	13	,	,	,
kim	14	14	either	either	CC
kim	14	15	with	with	IN
kim	14	16	its	-PRON-	PRP$
kim	14	17	carbon	carbon	NN
kim	14	18	-	-	HYPH
kim	14	19	based	base	VBN
kim	14	20	substrate	substrate	NN
kim	14	21	or	or	CC
kim	14	22	with	with	IN
kim	14	23	human	human	NN
kim	14	24	-	-	HYPH
kim	14	25	like	like	JJ
kim	14	26	understanding	understanding	NN
kim	14	27	that	that	WDT
kim	14	28	relies	rely	VBZ
kim	14	29	on	on	IN
kim	14	30	what	what	WP
kim	14	31	we	-PRON-	PRP
kim	14	32	regard	regard	VBP
kim	14	33	as	as	IN
kim	14	34	uniquely	uniquely	RB
kim	14	35	human	human	JJ
kim	14	36	cognitive	cognitive	JJ
kim	14	37	states	state	NNS
kim	14	38	such	such	JJ
kim	14	39	as	as	IN
kim	14	40	consciousness	consciousness	NN
kim	14	41	,	,	,
kim	14	42	qualia	quale	NNS
kim	14	43	,	,	,
kim	14	44	emotions	emotion	NNS
kim	14	45	,	,	,
kim	14	46	and	and	CC
kim	14	47	so	so	RB
kim	14	48	on	on	RB
kim	14	49	.	.	.
kim	15	1	For	for	IN
kim	15	2	more	more	RBR
kim	15	3	helpful	helpful	JJ
kim	15	4	descriptions	description	NNS
kim	15	5	of	of	IN
kim	15	6	common	common	JJ
kim	15	7	terms	term	NNS
kim	15	8	in	in	IN
kim	15	9	AI	AI	NNP
kim	15	10	,	,	,
kim	15	11	see	see	VB
kim	15	12	Tegmark	Tegmark	NNP
kim	15	13	2017	2017	CD
kim	15	14	,	,	,
kim	15	15	39	39	CD
kim	15	16	.	.	.
kim	16	1	For	for	IN
kim	16	2	more	more	JJR
kim	16	3	on	on	IN
kim	16	4	the	the	DT
kim	16	5	match	match	NN
kim	16	6	between	between	IN
kim	16	7	AlphaGo	AlphaGo	NNP
kim	16	8	and	and	CC
kim	16	9	Sedol	Sedol	NNP
kim	16	10	Lee	Lee	NNP
kim	16	11	,	,	,
kim	16	12	see	see	VB
kim	16	13	Koch	Koch	NNP
kim	16	14	2016	2016	CD
kim	16	15	.	.	.
kim	16	16	]	]	-RRB-
kim	17	1	[	[	-LRB-
kim	17	2	2	2	CD
kim	17	3	:	:	:
kim	17	4	Deep	deep	JJ
kim	17	5	reinforcement	reinforcement	NN
kim	17	6	learning	learning	NN
kim	17	7	is	be	VBZ
kim	17	8	a	a	DT
kim	17	9	type	type	NN
kim	17	10	of	of	IN
kim	17	11	deep	deep	JJ
kim	17	12	learning	learning	NN
kim	17	13	that	that	WDT
kim	17	14	is	be	VBZ
kim	17	15	goal	goal	NN
kim	17	16	-	-	HYPH
kim	17	17	oriented	orient	VBN
kim	17	18	and	and	CC
kim	17	19	reward	reward	NN
kim	17	20	-	-	HYPH
kim	17	21	based	base	VBN
kim	17	22	.	.	.
kim	18	1	See	see	VB
kim	18	2	Heaven	Heaven	NNP
kim	18	3	2020	2020	CD
kim	18	4	.	.	.
kim	18	5	]	]	-RRB-
kim	19	1	Early	early	JJ
kim	19	2	symbolic	symbolic	JJ
kim	19	3	AI	AI	NNP
kim	19	4	systems	system	NNS
kim	19	5	determined	determine	VBD
kim	19	6	their	-PRON-	PRP$
kim	19	7	outputs	output	NNS
kim	19	8	based	base	VBN
kim	19	9	upon	upon	IN
kim	19	10	given	give	VBN
kim	19	11	rules	rule	NNS
kim	19	12	and	and	CC
kim	19	13	logical	logical	JJ
kim	19	14	inference	inference	NN
kim	19	15	.	.	.
kim	20	1	AI	AI	NNP
kim	20	2	algorithms	algorithm	NNS
kim	20	3	in	in	IN
kim	20	4	these	these	DT
kim	20	5	rule	rule	NN
kim	20	6	-	-	HYPH
kim	20	7	based	base	VBN
kim	20	8	systems	system	NNS
kim	20	9	,	,	,
kim	20	10	also	also	RB
kim	20	11	known	know	VBN
kim	20	12	as	as	IN
kim	20	13	good	good	JJ
kim	20	14	old	old	JJ
kim	20	15	-	-	HYPH
kim	20	16	fashioned	fashioned	JJ
kim	20	17	AI	AI	NNP
kim	20	18	(	(	-LRB-
kim	20	19	GOFAI	GOFAI	NNP
kim	20	20	)	)	-RRB-
kim	20	21	,	,	,
kim	20	22	were	be	VBD
kim	20	23	pre	pre	VBN
kim	20	24	-	-	VBN
kim	20	25	determined	determined	JJ
kim	20	26	,	,	,
kim	20	27	predictable	predictable	JJ
kim	20	28	,	,	,
kim	20	29	and	and	CC
kim	20	30	transparent	transparent	JJ
kim	20	31	.	.	.
kim	21	1	On	on	IN
kim	21	2	the	the	DT
kim	21	3	other	other	JJ
kim	21	4	hand	hand	NN
kim	21	5	,	,	,
kim	21	6	machine	machine	NN
kim	21	7	learning	learning	NN
kim	21	8	,	,	,
kim	21	9	another	another	DT
kim	21	10	approach	approach	NN
kim	21	11	in	in	IN
kim	21	12	AI	AI	NNP
kim	21	13	,	,	,
kim	21	14	enables	enable	VBZ
kim	21	15	an	an	DT
kim	21	16	AI	ai	NN
kim	21	17	algorithm	algorithm	NN
kim	21	18	to	to	TO
kim	21	19	evolve	evolve	VB
kim	21	20	.	.	.
kim	22	1	It	-PRON-	PRP
kim	22	2	learns	learn	VBZ
kim	22	3	to	to	TO
kim	22	4	adjust	adjust	VB
kim	22	5	itself	-PRON-	PRP
kim	22	6	to	to	TO
kim	22	7	identify	identify	VB
kim	22	8	certain	certain	JJ
kim	22	9	patterns	pattern	NNS
kim	22	10	through	through	IN
kim	22	11	the	the	DT
kim	22	12	so	so	RB
kim	22	13	-	-	HYPH
kim	22	14	called	call	VBN
kim	22	15	‘	'	``
kim	22	16	training	training	NN
kim	22	17	’	'	''
kim	22	18	process	process	NN
kim	22	19	,	,	,
kim	22	20	which	which	WDT
kim	22	21	relies	rely	VBZ
kim	22	22	on	on	IN
kim	22	23	a	a	DT
kim	22	24	large	large	JJ
kim	22	25	amount	amount	NN
kim	22	26	of	of	IN
kim	22	27	data	datum	NNS
kim	22	28	and	and	CC
kim	22	29	statistics	statistic	NNS
kim	22	30	.	.	.
kim	23	1	Deep	deep	JJ
kim	23	2	learning	learning	NN
kim	23	3	,	,	,
kim	23	4	one	one	CD
kim	23	5	of	of	IN
kim	23	6	the	the	DT
kim	23	7	widely	widely	RB
kim	23	8	-	-	HYPH
kim	23	9	used	use	VBN
kim	23	10	techniques	technique	NNS
kim	23	11	in	in	IN
kim	23	12	machine	machine	NN
kim	23	13	learning	learning	NN
kim	23	14	,	,	,
kim	23	15	further	further	JJ
kim	23	16	refines	refine	NNS
kim	23	17	this	this	DT
kim	23	18	training	training	NN
kim	23	19	process	process	NN
kim	23	20	using	use	VBG
kim	23	21	a	a	DT
kim	23	22	‘	'	``
kim	23	23	neural	neural	JJ
kim	23	24	network	network	NN
kim	23	25	.	.	.
kim	23	26	’[footnoteRef:3	’[footnoteref:3	ADD
kim	23	27	]	]	-RRB-
kim	23	28	Machine	machine	NN
kim	23	29	learning	learning	NN
kim	23	30	and	and	CC
kim	23	31	deep	deep	JJ
kim	23	32	learning	learning	NN
kim	23	33	have	have	VBP
kim	23	34	brought	bring	VBN
kim	23	35	significant	significant	JJ
kim	23	36	improvements	improvement	NNS
kim	23	37	to	to	IN
kim	23	38	the	the	DT
kim	23	39	performance	performance	NN
kim	23	40	of	of	IN
kim	23	41	AI	AI	NNP
kim	23	42	systems	system	NNS
kim	23	43	in	in	IN
kim	23	44	areas	area	NNS
kim	23	45	such	such	JJ
kim	23	46	as	as	IN
kim	23	47	translation	translation	NN
kim	23	48	,	,	,
kim	23	49	speech	speech	NN
kim	23	50	recognition	recognition	NN
kim	23	51	,	,	,
kim	23	52	and	and	CC
kim	23	53	detecting	detect	VBG
kim	23	54	objects	object	NNS
kim	23	55	and	and	CC
kim	23	56	predicting	predict	VBG
kim	23	57	their	-PRON-	PRP$
kim	23	58	movements	movement	NNS
kim	23	59	.	.	.
kim	24	1	Some	some	DT
kim	24	2	people	people	NNS
kim	24	3	assume	assume	VBP
kim	24	4	that	that	IN
kim	24	5	machine	machine	NN
kim	24	6	learning	learn	VBG
kim	24	7	completely	completely	RB
kim	24	8	replaced	replace	VBD
kim	24	9	GOFAI	GOFAI	NNP
kim	24	10	,	,	,
kim	24	11	but	but	CC
kim	24	12	this	this	DT
kim	24	13	is	be	VBZ
kim	24	14	a	a	DT
kim	24	15	misunderstanding	misunderstanding	NN
kim	24	16	.	.	.
kim	25	1	Symbolic	symbolic	JJ
kim	25	2	reasoning	reasoning	NN
kim	25	3	and	and	CC
kim	25	4	machine	machine	NN
kim	25	5	learning	learning	NN
kim	25	6	are	be	VBP
kim	25	7	two	two	CD
kim	25	8	distinct	distinct	JJ
kim	25	9	but	but	CC
kim	25	10	not	not	RB
kim	25	11	mutually	mutually	RB
kim	25	12	exclusive	exclusive	JJ
kim	25	13	approaches	approach	NNS
kim	25	14	,	,	,
kim	25	15	and	and	CC
kim	25	16	they	-PRON-	PRP
kim	25	17	can	can	MD
kim	25	18	be	be	VB
kim	25	19	used	use	VBN
kim	25	20	together	together	RB
kim	25	21	(	(	-LRB-
kim	25	22	Knight	Knight	NNP
kim	25	23	2019a	2019a	CD
kim	25	24	)	)	-RRB-
kim	25	25	.	.	.
kim	26	1	[	[	-LRB-
kim	26	2	3	3	CD
kim	26	3	:	:	:
kim	26	4	Machine	machine	NN
kim	26	5	learning	learning	NN
kim	26	6	and	and	CC
kim	26	7	deep	deep	JJ
kim	26	8	learning	learning	NN
kim	26	9	have	have	VBP
kim	26	10	gained	gain	VBN
kim	26	11	momentum	momentum	NN
kim	26	12	as	as	IN
kim	26	13	the	the	DT
kim	26	14	cost	cost	NN
kim	26	15	of	of	IN
kim	26	16	high	high	JJ
kim	26	17	-	-	HYPH
kim	26	18	performance	performance	NN
kim	26	19	computing	computing	NN
kim	26	20	has	have	VBZ
kim	26	21	gone	go	VBN
kim	26	22	down	down	RB
kim	26	23	and	and	CC
kim	26	24	large	large	JJ
kim	26	25	data	data	NN
kim	26	26	sets	set	NNS
kim	26	27	have	have	VBP
kim	26	28	become	become	VBN
kim	26	29	available	available	JJ
kim	26	30	.	.	.
kim	27	1	For	for	IN
kim	27	2	example	example	NN
kim	27	3	,	,	,
kim	27	4	the	the	DT
kim	27	5	data	datum	NNS
kim	27	6	in	in	IN
kim	27	7	the	the	DT
kim	27	8	ImageNet	ImageNet	NNP
kim	27	9	contains	contain	VBZ
kim	27	10	more	more	JJR
kim	27	11	than	than	IN
kim	27	12	14	14	CD
kim	27	13	million	million	CD
kim	27	14	hand	hand	NN
kim	27	15	-	-	HYPH
kim	27	16	annotated	annotate	VBN
kim	27	17	images	image	NNS
kim	27	18	.	.	.
kim	28	1	The	the	DT
kim	28	2	ImageNet	ImageNet	NNP
kim	28	3	data	datum	NNS
kim	28	4	have	have	VBP
kim	28	5	been	be	VBN
kim	28	6	used	use	VBN
kim	28	7	for	for	IN
kim	28	8	the	the	DT
kim	28	9	well	well	RB
kim	28	10	-	-	HYPH
kim	28	11	known	know	VBN
kim	28	12	annual	annual	JJ
kim	28	13	AI	AI	NNP
kim	28	14	competition	competition	NN
kim	28	15	for	for	IN
kim	28	16	object	object	NN
kim	28	17	detection	detection	NN
kim	28	18	and	and	CC
kim	28	19	image	image	NN
kim	28	20	classification	classification	NN
kim	28	21	at	at	IN
kim	28	22	large	large	JJ
kim	28	23	scale	scale	NN
kim	28	24	from	from	IN
kim	28	25	2010	2010	CD
kim	28	26	to	to	IN
kim	28	27	2017	2017	CD
kim	28	28	.	.	.
kim	29	1	See	see	VB
kim	29	2	http://www.image-net.org/challenges/LSVRC/.	http://www.image-net.org/challenges/lsvrc/.	NN
kim	29	3	]	]	-RRB-
kim	30	1	With	with	IN
kim	30	2	their	-PRON-	PRP$
kim	30	3	limited	limit	VBN
kim	30	4	intelligence	intelligence	NN
kim	30	5	and	and	CC
kim	30	6	fully	fully	RB
kim	30	7	deterministic	deterministic	JJ
kim	30	8	nature	nature	NN
kim	30	9	,	,	,
kim	30	10	early	early	JJ
kim	30	11	rule	rule	NN
kim	30	12	-	-	HYPH
kim	30	13	based	base	VBN
kim	30	14	symbolic	symbolic	JJ
kim	30	15	AI	AI	NNP
kim	30	16	systems	system	NNS
kim	30	17	raised	raise	VBD
kim	30	18	few	few	JJ
kim	30	19	ethical	ethical	JJ
kim	30	20	concerns	concern	NNS
kim	30	21	.	.	.
kim	31	1	[	[	-LRB-
kim	31	2	footnoteRef:4	footnoteRef:4	NNS
kim	31	3	]	]	-RRB-
kim	31	4	AI	AI	NNP
kim	31	5	systems	system	NNS
kim	31	6	that	that	IN
kim	31	7	near	near	RB
kim	31	8	or	or	CC
kim	31	9	surpass	surpass	VB
kim	31	10	human	human	JJ
kim	31	11	capability	capability	NN
kim	31	12	,	,	,
kim	31	13	on	on	IN
kim	31	14	the	the	DT
kim	31	15	other	other	JJ
kim	31	16	hand	hand	NN
kim	31	17	,	,	,
kim	31	18	are	be	VBP
kim	31	19	likely	likely	JJ
kim	31	20	to	to	TO
kim	31	21	be	be	VB
kim	31	22	given	give	VBN
kim	31	23	the	the	DT
kim	31	24	autonomy	autonomy	NN
kim	31	25	to	to	TO
kim	31	26	make	make	VB
kim	31	27	their	-PRON-	PRP$
kim	31	28	own	own	JJ
kim	31	29	decisions	decision	NNS
kim	31	30	without	without	IN
kim	31	31	humans	human	NNS
kim	31	32	,	,	,
kim	31	33	even	even	RB
kim	31	34	when	when	WRB
kim	31	35	their	-PRON-	PRP$
kim	31	36	workings	working	NNS
kim	31	37	are	be	VBP
kim	31	38	not	not	RB
kim	31	39	entirely	entirely	RB
kim	31	40	transparent	transparent	JJ
kim	31	41	,	,	,
kim	31	42	and	and	CC
kim	31	43	some	some	DT
kim	31	44	of	of	IN
kim	31	45	those	those	DT
kim	31	46	decisions	decision	NNS
kim	31	47	are	be	VBP
kim	31	48	distinctively	distinctively	RB
kim	31	49	moral	moral	JJ
kim	31	50	in	in	IN
kim	31	51	character	character	NN
kim	31	52	.	.	.
kim	32	1	As	as	IN
kim	32	2	humans	human	NNS
kim	32	3	,	,	,
kim	32	4	we	-PRON-	PRP
kim	32	5	are	be	VBP
kim	32	6	trained	train	VBN
kim	32	7	to	to	TO
kim	32	8	recognize	recognize	VB
kim	32	9	situations	situation	NNS
kim	32	10	that	that	WDT
kim	32	11	demand	demand	VBP
kim	32	12	moral	moral	JJ
kim	32	13	decision	decision	NN
kim	32	14	-	-	HYPH
kim	32	15	making	making	NN
kim	32	16	.	.	.
kim	33	1	But	but	CC
kim	33	2	how	how	WRB
kim	33	3	would	would	MD
kim	33	4	an	an	DT
kim	33	5	AI	ai	NN
kim	33	6	system	system	NN
kim	33	7	be	be	VB
kim	33	8	able	able	JJ
kim	33	9	to	to	TO
kim	33	10	do	do	VB
kim	33	11	so	so	RB
kim	33	12	,	,	,
kim	33	13	and	and	CC
kim	33	14	should	should	MD
kim	33	15	they	-PRON-	PRP
kim	33	16	be	be	VB
kim	33	17	allowed	allow	VBN
kim	33	18	to	to	IN
kim	33	19	?	?	.
kim	34	1	With	with	IN
kim	34	2	self	self	NN
kim	34	3	-	-	HYPH
kim	34	4	driving	drive	VBG
kim	34	5	cars	car	NNS
kim	34	6	and	and	CC
kim	34	7	autonomous	autonomous	JJ
kim	34	8	weapons	weapon	NNS
kim	34	9	systems	system	NNS
kim	34	10	under	under	IN
kim	34	11	active	active	JJ
kim	34	12	development	development	NN
kim	34	13	and	and	CC
kim	34	14	testing	testing	NN
kim	34	15	,	,	,
kim	34	16	these	these	DT
kim	34	17	are	be	VBP
kim	34	18	no	no	RB
kim	34	19	longer	long	RBR
kim	34	20	idle	idle	JJ
kim	34	21	questions	question	NNS
kim	34	22	.	.	.
kim	35	1	[	[	-LRB-
kim	35	2	4	4	CD
kim	35	3	:	:	:
kim	35	4	For	for	IN
kim	35	5	an	an	DT
kim	35	6	excellent	excellent	JJ
kim	35	7	history	history	NN
kim	35	8	of	of	IN
kim	35	9	AI	AI	NNP
kim	35	10	research	research	NN
kim	35	11	,	,	,
kim	35	12	see	see	VB
kim	35	13	chapter	chapter	NN
kim	35	14	1	1	CD
kim	35	15	,	,	,
kim	35	16	“	"	``
kim	35	17	What	what	WP
kim	35	18	is	be	VBZ
kim	35	19	Artificial	Artificial	NNP
kim	35	20	Intelligence	Intelligence	NNP
kim	35	21	,	,	,
kim	35	22	”	"	''
kim	35	23	of	of	IN
kim	35	24	Boden	Boden	NNP
kim	35	25	2016	2016	CD
kim	35	26	,	,	,
kim	35	27	1	1	CD
kim	35	28	-	-	SYM
kim	35	29	20	20	CD
kim	35	30	.	.	.
kim	35	31	]	]	-RRB-
kim	36	1	2	2	LS
kim	36	2	.	.	.
kim	37	1	The	the	DT
kim	37	2	Trolley	Trolley	NNP
kim	37	3	Problem	Problem	NNP
kim	37	4	Recent	recent	JJ
kim	37	5	advances	advance	NNS
kim	37	6	of	of	IN
kim	37	7	AI	AI	NNP
kim	37	8	,	,	,
kim	37	9	such	such	JJ
kim	37	10	as	as	IN
kim	37	11	autonomous	autonomous	JJ
kim	37	12	cars	car	NNS
kim	37	13	,	,	,
kim	37	14	have	have	VBP
kim	37	15	brought	bring	VBN
kim	37	16	new	new	JJ
kim	37	17	interest	interest	NN
kim	37	18	to	to	IN
kim	37	19	the	the	DT
kim	37	20	trolley	trolley	NN
kim	37	21	problem	problem	NN
kim	37	22	,	,	,
kim	37	23	a	a	DT
kim	37	24	thought	thought	NN
kim	37	25	experiment	experiment	NN
kim	37	26	introduced	introduce	VBN
kim	37	27	by	by	IN
kim	37	28	Philippa	Philippa	NNP
kim	37	29	Foot	Foot	NNP
kim	37	30	in	in	IN
kim	37	31	1967	1967	CD
kim	37	32	.	.	.
kim	38	1	In	in	IN
kim	38	2	the	the	DT
kim	38	3	standard	standard	JJ
kim	38	4	version	version	NN
kim	38	5	of	of	IN
kim	38	6	this	this	DT
kim	38	7	problem	problem	NN
kim	38	8	,	,	,
kim	38	9	a	a	DT
kim	38	10	runaway	runaway	JJ
kim	38	11	trolley	trolley	NN
kim	38	12	barrels	barrel	NNS
kim	38	13	down	down	IN
kim	38	14	a	a	DT
kim	38	15	track	track	NN
kim	38	16	where	where	WRB
kim	38	17	five	five	CD
kim	38	18	unsuspecting	unsuspecting	JJ
kim	38	19	people	people	NNS
kim	38	20	are	be	VBP
kim	38	21	standing	stand	VBG
kim	38	22	.	.	.
kim	39	1	You	-PRON-	PRP
kim	39	2	happen	happen	VBP
kim	39	3	to	to	TO
kim	39	4	be	be	VB
kim	39	5	standing	stand	VBG
kim	39	6	next	next	RB
kim	39	7	to	to	IN
kim	39	8	a	a	DT
kim	39	9	lever	lever	NN
kim	39	10	that	that	WDT
kim	39	11	switches	switch	VBZ
kim	39	12	the	the	DT
kim	39	13	trolley	trolley	NN
kim	39	14	onto	onto	IN
kim	39	15	a	a	DT
kim	39	16	different	different	JJ
kim	39	17	track	track	NN
kim	39	18	,	,	,
kim	39	19	where	where	WRB
kim	39	20	there	there	EX
kim	39	21	is	be	VBZ
kim	39	22	only	only	RB
kim	39	23	one	one	CD
kim	39	24	person	person	NN
kim	39	25	.	.	.
kim	40	1	Those	those	DT
kim	40	2	who	who	WP
kim	40	3	are	be	VBP
kim	40	4	on	on	IN
kim	40	5	either	either	DT
kim	40	6	track	track	NN
kim	40	7	will	will	MD
kim	40	8	be	be	VB
kim	40	9	killed	kill	VBN
kim	40	10	if	if	IN
kim	40	11	the	the	DT
kim	40	12	trolley	trolley	NN
kim	40	13	heads	head	VBZ
kim	40	14	their	-PRON-	PRP$
kim	40	15	way	way	NN
kim	40	16	.	.	.
kim	41	1	Should	Should	MD
kim	41	2	you	-PRON-	PRP
kim	41	3	pull	pull	VB
kim	41	4	the	the	DT
kim	41	5	lever	lever	NN
kim	41	6	,	,	,
kim	41	7	so	so	IN
kim	41	8	that	that	IN
kim	41	9	the	the	DT
kim	41	10	runaway	runaway	JJ
kim	41	11	trolley	trolley	NN
kim	41	12	would	would	MD
kim	41	13	kill	kill	VB
kim	41	14	one	one	CD
kim	41	15	person	person	NN
kim	41	16	instead	instead	RB
kim	41	17	of	of	IN
kim	41	18	five	five	CD
kim	41	19	?	?	.
kim	42	1	Unlike	unlike	IN
kim	42	2	people	people	NNS
kim	42	3	,	,	,
kim	42	4	a	a	DT
kim	42	5	machine	machine	NN
kim	42	6	does	do	VBZ
kim	42	7	not	not	RB
kim	42	8	panic	panic	VB
kim	42	9	or	or	CC
kim	42	10	freeze	freeze	VB
kim	42	11	and	and	CC
kim	42	12	simply	simply	RB
kim	42	13	follows	follow	VBZ
kim	42	14	and	and	CC
kim	42	15	executes	execute	VBZ
kim	42	16	the	the	DT
kim	42	17	given	give	VBN
kim	42	18	instruction	instruction	NN
kim	42	19	.	.	.
kim	43	1	So	so	RB
kim	43	2	,	,	,
kim	43	3	this	this	DT
kim	43	4	opens	open	VBZ
kim	43	5	up	up	RP
kim	43	6	the	the	DT
kim	43	7	possibility	possibility	NN
kim	43	8	of	of	IN
kim	43	9	building	build	VBG
kim	43	10	an	an	DT
kim	43	11	AI	AI	NNP
kim	43	12	-	-	HYPH
kim	43	13	powered	power	VBN
kim	43	14	trolley	trolley	NN
kim	43	15	that	that	WDT
kim	43	16	can	can	MD
kim	43	17	act	act	VB
kim	43	18	morally	morally	RB
kim	43	19	.	.	.
kim	44	1	[	[	-LRB-
kim	44	2	footnoteRef:5	footnoteRef:5	NNP
kim	44	3	]	]	-RRB-
kim	44	4	The	the	DT
kim	44	5	question	question	NN
kim	44	6	itself	-PRON-	PRP
kim	44	7	remains	remain	VBZ
kim	44	8	,	,	,
kim	44	9	however	however	RB
kim	44	10	.	.	.
kim	45	1	Should	Should	MD
kim	45	2	one	one	CD
kim	45	3	program	program	NN
kim	45	4	the	the	DT
kim	45	5	AI	ai	NN
kim	45	6	-	-	HYPH
kim	45	7	powered	power	VBN
kim	45	8	trolley	trolley	NN
kim	45	9	to	to	TO
kim	45	10	swerve	swerve	VB
kim	45	11	or	or	CC
kim	45	12	stay	stay	VB
kim	45	13	on	on	RP
kim	45	14	course	course	NN
kim	45	15	?	?	.
kim	46	1	[	[	-LRB-
kim	46	2	5	5	CD
kim	46	3	:	:	:
kim	46	4	Programming	programming	NN
kim	46	5	here	here	RB
kim	46	6	does	do	VBZ
kim	46	7	not	not	RB
kim	46	8	exclusively	exclusively	RB
kim	46	9	mean	mean	VB
kim	46	10	a	a	DT
kim	46	11	deep	deep	JJ
kim	46	12	learning	learning	NN
kim	46	13	or	or	CC
kim	46	14	machine	machine	NN
kim	46	15	learning	learning	NN
kim	46	16	approach	approach	NN
kim	46	17	.	.	.
kim	46	18	]	]	-RRB-
kim	47	1	Different	different	JJ
kim	47	2	moral	moral	JJ
kim	47	3	theories	theory	NNS
kim	47	4	,	,	,
kim	47	5	such	such	JJ
kim	47	6	as	as	IN
kim	47	7	virtue	virtue	NN
kim	47	8	ethics	ethic	NNS
kim	47	9	,	,	,
kim	47	10	contractarianism	contractarianism	NN
kim	47	11	,	,	,
kim	47	12	and	and	CC
kim	47	13	moral	moral	JJ
kim	47	14	relativism	relativism	NN
kim	47	15	,	,	,
kim	47	16	take	take	VBP
kim	47	17	different	different	JJ
kim	47	18	positions	position	NNS
kim	47	19	.	.	.
kim	48	1	Here	here	RB
kim	48	2	,	,	,
kim	48	3	I	-PRON-	PRP
kim	48	4	will	will	MD
kim	48	5	consider	consider	VB
kim	48	6	utilitarianism	utilitarianism	NN
kim	48	7	and	and	CC
kim	48	8	deontology	deontology	NN
kim	48	9	since	since	IN
kim	48	10	their	-PRON-	PRP$
kim	48	11	tenets	tenet	NNS
kim	48	12	are	be	VBP
kim	48	13	relatively	relatively	RB
kim	48	14	straightforward	straightforward	JJ
kim	48	15	and	and	CC
kim	48	16	most	most	JJS
kim	48	17	AI	AI	NNP
kim	48	18	developers	developer	NNS
kim	48	19	are	be	VBP
kim	48	20	likely	likely	JJ
kim	48	21	to	to	TO
kim	48	22	look	look	VB
kim	48	23	towards	towards	IN
kim	48	24	those	those	DT
kim	48	25	two	two	CD
kim	48	26	moral	moral	JJ
kim	48	27	theories	theory	NNS
kim	48	28	for	for	IN
kim	48	29	guidance	guidance	NN
kim	48	30	and	and	CC
kim	48	31	insight	insight	NN
kim	48	32	.	.	.
kim	49	1	Utilitarianism	Utilitarianism	NNP
kim	49	2	argues	argue	VBZ
kim	49	3	that	that	IN
kim	49	4	the	the	DT
kim	49	5	utility	utility	NN
kim	49	6	of	of	IN
kim	49	7	an	an	DT
kim	49	8	action	action	NN
kim	49	9	is	be	VBZ
kim	49	10	what	what	WP
kim	49	11	makes	make	VBZ
kim	49	12	an	an	DT
kim	49	13	action	action	NN
kim	49	14	moral	moral	JJ
kim	49	15	.	.	.
kim	50	1	In	in	IN
kim	50	2	this	this	DT
kim	50	3	view	view	NN
kim	50	4	,	,	,
kim	50	5	what	what	WP
kim	50	6	generates	generate	VBZ
kim	50	7	the	the	DT
kim	50	8	greatest	great	JJS
kim	50	9	amount	amount	NN
kim	50	10	of	of	IN
kim	50	11	good	good	JJ
kim	50	12	is	be	VBZ
kim	50	13	the	the	DT
kim	50	14	most	most	RBS
kim	50	15	moral	moral	JJ
kim	50	16	thing	thing	NN
kim	50	17	to	to	TO
kim	50	18	do	do	VB
kim	50	19	.	.	.
kim	51	1	If	if	IN
kim	51	2	one	one	PRP
kim	51	3	regards	regard	VBZ
kim	51	4	five	five	CD
kim	51	5	human	human	JJ
kim	51	6	lives	life	NNS
kim	51	7	as	as	IN
kim	51	8	a	a	DT
kim	51	9	greater	great	JJR
kim	51	10	good	good	NN
kim	51	11	than	than	IN
kim	51	12	one	one	CD
kim	51	13	,	,	,
kim	51	14	then	then	RB
kim	51	15	one	one	CD
kim	51	16	acts	act	VBZ
kim	51	17	morally	morally	RB
kim	51	18	by	by	IN
kim	51	19	pulling	pull	VBG
kim	51	20	the	the	DT
kim	51	21	lever	lever	NN
kim	51	22	and	and	CC
kim	51	23	diverting	divert	VBG
kim	51	24	the	the	DT
kim	51	25	trolley	trolley	NN
kim	51	26	to	to	IN
kim	51	27	the	the	DT
kim	51	28	other	other	JJ
kim	51	29	track	track	NN
kim	51	30	.	.	.
kim	52	1	By	by	IN
kim	52	2	contrast	contrast	NN
kim	52	3	,	,	,
kim	52	4	deontology	deontology	NN
kim	52	5	claims	claim	VBZ
kim	52	6	that	that	IN
kim	52	7	what	what	WP
kim	52	8	determines	determine	VBZ
kim	52	9	whether	whether	IN
kim	52	10	an	an	DT
kim	52	11	action	action	NN
kim	52	12	is	be	VBZ
kim	52	13	morally	morally	RB
kim	52	14	right	right	JJ
kim	52	15	or	or	CC
kim	52	16	wrong	wrong	JJ
kim	52	17	is	be	VBZ
kim	52	18	not	not	RB
kim	52	19	its	-PRON-	PRP$
kim	52	20	utility	utility	NN
kim	52	21	but	but	CC
kim	52	22	moral	moral	JJ
kim	52	23	rules	rule	NNS
kim	52	24	.	.	.
kim	53	1	If	if	IN
kim	53	2	an	an	DT
kim	53	3	action	action	NN
kim	53	4	is	be	VBZ
kim	53	5	in	in	IN
kim	53	6	accordance	accordance	NN
kim	53	7	with	with	IN
kim	53	8	those	those	DT
kim	53	9	rules	rule	NNS
kim	53	10	,	,	,
kim	53	11	then	then	RB
kim	53	12	the	the	DT
kim	53	13	action	action	NN
kim	53	14	is	be	VBZ
kim	53	15	morally	morally	RB
kim	53	16	right	right	JJ
kim	53	17	.	.	.
kim	54	1	Otherwise	otherwise	RB
kim	54	2	,	,	,
kim	54	3	it	-PRON-	PRP
kim	54	4	is	be	VBZ
kim	54	5	morally	morally	RB
kim	54	6	wrong	wrong	JJ
kim	54	7	.	.	.
kim	55	1	If	if	IN
kim	55	2	not	not	RB
kim	55	3	to	to	TO
kim	55	4	kill	kill	VB
kim	55	5	another	another	DT
kim	55	6	human	human	NN
kim	55	7	being	being	NN
kim	55	8	is	be	VBZ
kim	55	9	one	one	CD
kim	55	10	of	of	IN
kim	55	11	those	those	DT
kim	55	12	moral	moral	JJ
kim	55	13	rules	rule	NNS
kim	55	14	,	,	,
kim	55	15	then	then	RB
kim	55	16	killing	kill	VBG
kim	55	17	someone	someone	NN
kim	55	18	is	be	VBZ
kim	55	19	morally	morally	RB
kim	55	20	wrong	wrong	JJ
kim	55	21	even	even	RB
kim	55	22	if	if	IN
kim	55	23	it	-PRON-	PRP
kim	55	24	is	be	VBZ
kim	55	25	to	to	TO
kim	55	26	save	save	VB
kim	55	27	more	more	JJR
kim	55	28	lives	life	NNS
kim	55	29	.	.	.
kim	56	1	Note	note	VB
kim	56	2	that	that	IN
kim	56	3	these	these	DT
kim	56	4	are	be	VBP
kim	56	5	highly	highly	RB
kim	56	6	simplified	simplified	JJ
kim	56	7	accounts	account	NNS
kim	56	8	of	of	IN
kim	56	9	utilitarianism	utilitarianism	NN
kim	56	10	and	and	CC
kim	56	11	deontology	deontology	NN
kim	56	12	.	.	.
kim	57	1	The	the	DT
kim	57	2	good	good	NN
kim	57	3	in	in	IN
kim	57	4	utilitarianism	utilitarianism	NN
kim	57	5	can	can	MD
kim	57	6	be	be	VB
kim	57	7	interpreted	interpret	VBN
kim	57	8	in	in	IN
kim	57	9	many	many	JJ
kim	57	10	different	different	JJ
kim	57	11	ways	way	NNS
kim	57	12	,	,	,
kim	57	13	and	and	CC
kim	57	14	the	the	DT
kim	57	15	issue	issue	NN
kim	57	16	of	of	IN
kim	57	17	conflicting	conflict	VBG
kim	57	18	moral	moral	JJ
kim	57	19	rules	rule	NNS
kim	57	20	is	be	VBZ
kim	57	21	a	a	DT
kim	57	22	perennial	perennial	JJ
kim	57	23	problem	problem	NN
kim	57	24	that	that	IN
kim	57	25	deontological	deontological	JJ
kim	57	26	ethics	ethic	NNS
kim	57	27	grapples	grapple	VBZ
kim	57	28	with	with	IN
kim	57	29	.	.	.
kim	58	1	[	[	-LRB-
kim	58	2	footnoteRef:6	footnoteref:6	NN
kim	58	3	]	]	-RRB-
kim	58	4	For	for	IN
kim	58	5	our	-PRON-	PRP$
kim	58	6	purpose	purpose	NN
kim	58	7	,	,	,
kim	58	8	however	however	RB
kim	58	9	,	,	,
kim	58	10	these	these	DT
kim	58	11	simplified	simplify	VBN
kim	58	12	accounts	account	NNS
kim	58	13	are	be	VBP
kim	58	14	sufficient	sufficient	JJ
kim	58	15	to	to	TO
kim	58	16	highlight	highlight	VB
kim	58	17	the	the	DT
kim	58	18	aspects	aspect	NNS
kim	58	19	in	in	IN
kim	58	20	which	which	WDT
kim	58	21	the	the	DT
kim	58	22	utilitarian	utilitarian	NN
kim	58	23	and	and	CC
kim	58	24	the	the	DT
kim	58	25	deontological	deontological	JJ
kim	58	26	position	position	NN
kim	58	27	appeal	appeal	VBP
kim	58	28	to	to	IN
kim	58	29	and	and	CC
kim	58	30	go	go	VB
kim	58	31	against	against	IN
kim	58	32	our	-PRON-	PRP$
kim	58	33	moral	moral	JJ
kim	58	34	intuition	intuition	NN
kim	58	35	.	.	.
kim	59	1	[	[	-LRB-
kim	59	2	6	6	CD
kim	59	3	:	:	:
kim	59	4	For	for	IN
kim	59	5	an	an	DT
kim	59	6	overview	overview	NN
kim	59	7	,	,	,
kim	59	8	see	see	VB
kim	59	9	Sinnott	Sinnott	NNP
kim	59	10	-	-	HYPH
kim	59	11	Armstrong	Armstrong	NNP
kim	59	12	2019	2019	CD
kim	59	13	and	and	CC
kim	59	14	Alexander	Alexander	NNP
kim	59	15	and	and	CC
kim	59	16	Moore	Moore	NNP
kim	59	17	2016	2016	CD
kim	59	18	.	.	.
kim	59	19	]	]	-RRB-
kim	60	1	If	if	IN
kim	60	2	a	a	DT
kim	60	3	trolley	trolley	NN
kim	60	4	can	can	MD
kim	60	5	not	not	RB
kim	60	6	be	be	VB
kim	60	7	stopped	stop	VBN
kim	60	8	,	,	,
kim	60	9	saving	save	VBG
kim	60	10	five	five	CD
kim	60	11	lives	life	NNS
kim	60	12	over	over	IN
kim	60	13	one	one	NN
kim	60	14	seems	seem	VBZ
kim	60	15	to	to	TO
kim	60	16	be	be	VB
kim	60	17	a	a	DT
kim	60	18	right	right	JJ
kim	60	19	thing	thing	NN
kim	60	20	to	to	TO
kim	60	21	do	do	VB
kim	60	22	.	.	.
kim	61	1	Utilitarianism	Utilitarianism	NNP
kim	61	2	appears	appear	VBZ
kim	61	3	to	to	TO
kim	61	4	get	get	VB
kim	61	5	things	thing	NNS
kim	61	6	right	right	JJ
kim	61	7	in	in	IN
kim	61	8	this	this	DT
kim	61	9	respect	respect	NN
kim	61	10	.	.	.
kim	62	1	However	however	RB
kim	62	2	,	,	,
kim	62	3	it	-PRON-	PRP
kim	62	4	is	be	VBZ
kim	62	5	hard	hard	JJ
kim	62	6	to	to	TO
kim	62	7	dispute	dispute	VB
kim	62	8	that	that	IN
kim	62	9	killing	kill	VBG
kim	62	10	people	people	NNS
kim	62	11	is	be	VBZ
kim	62	12	wrong	wrong	JJ
kim	62	13	.	.	.
kim	63	1	If	if	IN
kim	63	2	killing	killing	NN
kim	63	3	is	be	VBZ
kim	63	4	morally	morally	RB
kim	63	5	wrong	wrong	JJ
kim	63	6	no	no	RB
kim	63	7	matter	matter	RB
kim	63	8	what	what	WP
kim	63	9	,	,	,
kim	63	10	deontology	deontology	NN
kim	63	11	seems	seem	VBZ
kim	63	12	to	to	TO
kim	63	13	make	make	VB
kim	63	14	more	more	JJR
kim	63	15	sense	sense	NN
kim	63	16	.	.	.
kim	64	1	With	with	IN
kim	64	2	moral	moral	JJ
kim	64	3	theories	theory	NNS
kim	64	4	,	,	,
kim	64	5	things	thing	NNS
kim	64	6	seem	seem	VBP
kim	64	7	to	to	TO
kim	64	8	get	get	VB
kim	64	9	more	more	RBR
kim	64	10	confusing	confusing	JJ
kim	64	11	.	.	.
kim	65	1	Consider	consider	VB
kim	65	2	the	the	DT
kim	65	3	case	case	NN
kim	65	4	in	in	IN
kim	65	5	which	which	WDT
kim	65	6	one	one	NN
kim	65	7	freezes	freeze	VBZ
kim	65	8	and	and	CC
kim	65	9	fails	fail	VBZ
kim	65	10	to	to	TO
kim	65	11	pull	pull	VB
kim	65	12	the	the	DT
kim	65	13	level	level	NN
kim	65	14	.	.	.
kim	66	1	According	accord	VBG
kim	66	2	to	to	IN
kim	66	3	utilitarianism	utilitarianism	NN
kim	66	4	,	,	,
kim	66	5	this	this	DT
kim	66	6	would	would	MD
kim	66	7	be	be	VB
kim	66	8	morally	morally	RB
kim	66	9	wrong	wrong	JJ
kim	66	10	because	because	IN
kim	66	11	it	-PRON-	PRP
kim	66	12	fails	fail	VBZ
kim	66	13	to	to	TO
kim	66	14	maximize	maximize	VB
kim	66	15	the	the	DT
kim	66	16	greatest	great	JJS
kim	66	17	good	good	JJ
kim	66	18	,	,	,
kim	66	19	i.e.	i.e.	FW
kim	67	1	human	human	JJ
kim	67	2	lives	life	NNS
kim	67	3	.	.	.
kim	68	1	But	but	CC
kim	68	2	how	how	WRB
kim	68	3	far	far	RB
kim	68	4	should	should	MD
kim	68	5	one	one	PRP
kim	68	6	go	go	VB
kim	68	7	to	to	TO
kim	68	8	maximize	maximize	VB
kim	68	9	the	the	DT
kim	68	10	good	good	NN
kim	68	11	?	?	.
kim	69	1	Suppose	suppose	VB
kim	69	2	there	there	EX
kim	69	3	is	be	VBZ
kim	69	4	a	a	DT
kim	69	5	very	very	RB
kim	69	6	large	large	JJ
kim	69	7	person	person	NN
kim	69	8	on	on	IN
kim	69	9	a	a	DT
kim	69	10	footbridge	footbridge	NN
kim	69	11	over	over	IN
kim	69	12	the	the	DT
kim	69	13	trolley	trolley	NN
kim	69	14	track	track	NN
kim	69	15	and	and	CC
kim	69	16	one	one	NN
kim	69	17	pushes	push	VBZ
kim	69	18	that	that	DT
kim	69	19	person	person	NN
kim	69	20	off	off	IN
kim	69	21	the	the	DT
kim	69	22	footbridge	footbridge	NN
kim	69	23	onto	onto	IN
kim	69	24	the	the	DT
kim	69	25	track	track	NN
kim	69	26	,	,	,
kim	69	27	thus	thus	RB
kim	69	28	stopping	stop	VBG
kim	69	29	the	the	DT
kim	69	30	trolley	trolley	NN
kim	69	31	and	and	CC
kim	69	32	saving	save	VBG
kim	69	33	the	the	DT
kim	69	34	five	five	CD
kim	69	35	people	people	NNS
kim	69	36	.	.	.
kim	70	1	Would	Would	MD
kim	70	2	this	this	DT
kim	70	3	count	count	VB
kim	70	4	as	as	IN
kim	70	5	a	a	DT
kim	70	6	right	right	JJ
kim	70	7	thing	thing	NN
kim	70	8	to	to	TO
kim	70	9	do	do	VB
kim	70	10	?	?	.
kim	71	1	Utilitarianism	Utilitarianism	NNP
kim	71	2	may	may	MD
kim	71	3	argue	argue	VB
kim	71	4	that	that	DT
kim	71	5	.	.	.
kim	72	1	But	but	CC
kim	72	2	in	in	IN
kim	72	3	real	real	JJ
kim	72	4	life	life	NN
kim	72	5	,	,	,
kim	72	6	many	many	JJ
kim	72	7	might	may	MD
kim	72	8	consider	consider	VB
kim	72	9	throwing	throw	VBG
kim	72	10	a	a	DT
kim	72	11	person	person	NN
kim	72	12	morally	morally	RB
kim	72	13	wrong	wrong	JJ
kim	72	14	but	but	CC
kim	72	15	pulling	pull	VBG
kim	72	16	the	the	DT
kim	72	17	lever	lever	NN
kim	72	18	morally	morally	RB
kim	72	19	permissible	permissible	JJ
kim	72	20	.	.	.
kim	73	1	[	[	-LRB-
kim	73	2	footnoteRef:7	footnoteref:7	NN
kim	73	3	]	]	-RRB-
kim	73	4	[	[	-LRB-
kim	73	5	7	7	CD
kim	73	6	:	:	:
kim	73	7	For	for	IN
kim	73	8	an	an	DT
kim	73	9	empirical	empirical	JJ
kim	73	10	study	study	NN
kim	73	11	on	on	IN
kim	73	12	this	this	DT
kim	73	13	,	,	,
kim	73	14	see	see	VB
kim	73	15	Cushman	Cushman	NNP
kim	73	16	,	,	,
kim	73	17	Young	Young	NNP
kim	73	18	,	,	,
kim	73	19	and	and	CC
kim	73	20	Hauser	Hauser	NNP
kim	73	21	2006	2006	CD
kim	73	22	.	.	.
kim	74	1	For	for	IN
kim	74	2	the	the	DT
kim	74	3	results	result	NNS
kim	74	4	of	of	IN
kim	74	5	a	a	DT
kim	74	6	similar	similar	JJ
kim	74	7	survey	survey	NN
kim	74	8	that	that	WDT
kim	74	9	involves	involve	VBZ
kim	74	10	an	an	DT
kim	74	11	autonomous	autonomous	JJ
kim	74	12	car	car	NN
kim	74	13	instead	instead	RB
kim	74	14	of	of	IN
kim	74	15	a	a	DT
kim	74	16	trolley	trolley	NN
kim	74	17	,	,	,
kim	74	18	see	see	VB
kim	74	19	Bonnefon	Bonnefon	NNP
kim	74	20	,	,	,
kim	74	21	Shariff	Shariff	NNP
kim	74	22	,	,	,
kim	74	23	and	and	CC
kim	74	24	Rahwan	Rahwan	NNP
kim	74	25	2016	2016	CD
kim	74	26	.	.	.
kim	74	27	]	]	-RRB-
kim	75	1	The	the	DT
kim	75	2	problem	problem	NN
kim	75	3	with	with	IN
kim	75	4	utilitarianism	utilitarianism	NN
kim	75	5	is	be	VBZ
kim	75	6	that	that	IN
kim	75	7	it	-PRON-	PRP
kim	75	8	treats	treat	VBZ
kim	75	9	the	the	DT
kim	75	10	good	good	JJ
kim	75	11	as	as	IN
kim	75	12	something	something	NN
kim	75	13	inherently	inherently	RB
kim	75	14	quantifiable	quantifiable	JJ
kim	75	15	,	,	,
kim	75	16	comparable	comparable	JJ
kim	75	17	,	,	,
kim	75	18	calculable	calculable	JJ
kim	75	19	,	,	,
kim	75	20	and	and	CC
kim	75	21	additive	additive	NNP
kim	75	22	.	.	.
kim	76	1	But	but	CC
kim	76	2	not	not	RB
kim	76	3	all	all	DT
kim	76	4	considerations	consideration	NNS
kim	76	5	that	that	WDT
kim	76	6	we	-PRON-	PRP
kim	76	7	have	have	VBP
kim	76	8	to	to	TO
kim	76	9	factor	factor	VB
kim	76	10	into	into	IN
kim	76	11	moral	moral	JJ
kim	76	12	decision	decision	NN
kim	76	13	-	-	HYPH
kim	76	14	making	making	NN
kim	76	15	are	be	VBP
kim	76	16	measurable	measurable	JJ
kim	76	17	in	in	IN
kim	76	18	numbers	number	NNS
kim	76	19	.	.	.
kim	77	1	What	what	WP
kim	77	2	if	if	IN
kim	77	3	the	the	DT
kim	77	4	five	five	CD
kim	77	5	people	people	NNS
kim	77	6	on	on	IN
kim	77	7	the	the	DT
kim	77	8	track	track	NN
kim	77	9	are	be	VBP
kim	77	10	helpless	helpless	JJ
kim	77	11	babies	baby	NNS
kim	77	12	or	or	CC
kim	77	13	murderers	murderer	NNS
kim	77	14	who	who	WP
kim	77	15	just	just	RB
kim	77	16	escaped	escape	VBD
kim	77	17	from	from	IN
kim	77	18	the	the	DT
kim	77	19	prison	prison	NN
kim	77	20	?	?	.
kim	78	1	Would	Would	MD
kim	78	2	or	or	CC
kim	78	3	should	should	MD
kim	78	4	that	that	DT
kim	78	5	affect	affect	VB
kim	78	6	our	-PRON-	PRP$
kim	78	7	decision	decision	NN
kim	78	8	?	?	.
kim	79	1	Some	some	DT
kim	79	2	of	of	IN
kim	79	3	us	-PRON-	PRP
kim	79	4	would	would	MD
kim	79	5	surely	surely	RB
kim	79	6	hesitate	hesitate	VB
kim	79	7	saving	save	VBG
kim	79	8	the	the	DT
kim	79	9	lives	life	NNS
kim	79	10	of	of	IN
kim	79	11	five	five	CD
kim	79	12	murderers	murderer	NNS
kim	79	13	by	by	IN
kim	79	14	sacrificing	sacrifice	VBG
kim	79	15	one	one	CD
kim	79	16	innocent	innocent	JJ
kim	79	17	baby	baby	NN
kim	79	18	.	.	.
kim	80	1	But	but	CC
kim	80	2	what	what	WP
kim	80	3	if	if	IN
kim	80	4	things	thing	NNS
kim	80	5	were	be	VBD
kim	80	6	different	different	JJ
kim	80	7	and	and	CC
kim	80	8	we	-PRON-	PRP
kim	80	9	were	be	VBD
kim	80	10	comparing	compare	VBG
kim	80	11	five	five	CD
kim	80	12	school	school	NN
kim	80	13	children	child	NNS
kim	80	14	versus	versus	IN
kim	80	15	one	one	CD
kim	80	16	baby	baby	NN
kim	80	17	or	or	CC
kim	80	18	five	five	CD
kim	80	19	babies	baby	NNS
kim	80	20	versus	versus	IN
kim	80	21	one	one	CD
kim	80	22	schoolchild	schoolchild	NN
kim	80	23	?	?	.
kim	81	1	No	no	DT
kim	81	2	one	one	NN
kim	81	3	can	can	MD
kim	81	4	say	say	VB
kim	81	5	for	for	IN
kim	81	6	sure	sure	JJ
kim	81	7	what	what	WP
kim	81	8	is	be	VBZ
kim	81	9	the	the	DT
kim	81	10	morally	morally	RB
kim	81	11	right	right	JJ
kim	81	12	action	action	NN
kim	81	13	in	in	IN
kim	81	14	those	those	DT
kim	81	15	cases	case	NNS
kim	81	16	.	.	.
kim	82	1	[	[	-LRB-
kim	82	2	footnoteRef:8	footnoteref:8	NN
kim	82	3	]	]	-RRB-
kim	82	4	[	[	-LRB-
kim	82	5	8	8	CD
kim	82	6	:	:	:
kim	82	7	For	for	IN
kim	82	8	an	an	DT
kim	82	9	attempt	attempt	NN
kim	82	10	to	to	TO
kim	82	11	identify	identify	VB
kim	82	12	moral	moral	JJ
kim	82	13	principles	principle	NNS
kim	82	14	behind	behind	IN
kim	82	15	our	-PRON-	PRP$
kim	82	16	moral	moral	JJ
kim	82	17	intuition	intuition	NN
kim	82	18	in	in	IN
kim	82	19	different	different	JJ
kim	82	20	versions	version	NNS
kim	82	21	of	of	IN
kim	82	22	the	the	DT
kim	82	23	trolley	trolley	NN
kim	82	24	problem	problem	NN
kim	82	25	and	and	CC
kim	82	26	other	other	JJ
kim	82	27	similar	similar	JJ
kim	82	28	cases	case	NNS
kim	82	29	,	,	,
kim	82	30	see	see	VB
kim	82	31	Thomson	Thomson	NNP
kim	82	32	1976	1976	CD
kim	82	33	.	.	.
kim	82	34	]	]	-RRB-
kim	83	1	While	while	IN
kim	83	2	the	the	DT
kim	83	3	utilitarian	utilitarian	JJ
kim	83	4	position	position	NN
kim	83	5	appears	appear	VBZ
kim	83	6	less	less	RBR
kim	83	7	persuasive	persuasive	JJ
kim	83	8	in	in	IN
kim	83	9	light	light	NN
kim	83	10	of	of	IN
kim	83	11	these	these	DT
kim	83	12	considerations	consideration	NNS
kim	83	13	,	,	,
kim	83	14	deontology	deontology	NN
kim	83	15	does	do	VBZ
kim	83	16	n’t	not	RB
kim	83	17	fare	fare	VB
kim	83	18	too	too	RB
kim	83	19	well	well	RB
kim	83	20	,	,	,
kim	83	21	either	either	RB
kim	83	22	.	.	.
kim	84	1	Deontology	deontology	NN
kim	84	2	emphasizes	emphasize	VBZ
kim	84	3	one	one	PRP
kim	84	4	’s	’s	NN
kim	84	5	duty	duty	NN
kim	84	6	to	to	TO
kim	84	7	observe	observe	VB
kim	84	8	moral	moral	JJ
kim	84	9	rules	rule	NNS
kim	84	10	.	.	.
kim	85	1	But	but	CC
kim	85	2	what	what	WP
kim	85	3	if	if	IN
kim	85	4	those	those	DT
kim	85	5	moral	moral	JJ
kim	85	6	rules	rule	NNS
kim	85	7	conflict	conflict	NN
kim	85	8	with	with	IN
kim	85	9	one	one	CD
kim	85	10	another	another	DT
kim	85	11	?	?	.
kim	86	1	Between	between	IN
kim	86	2	the	the	DT
kim	86	3	two	two	CD
kim	86	4	moral	moral	JJ
kim	86	5	rules	rule	NNS
kim	86	6	,	,	,
kim	86	7	“	"	``
kim	86	8	do	do	VB
kim	86	9	not	not	RB
kim	86	10	kill	kill	VB
kim	86	11	a	a	DT
kim	86	12	person	person	NN
kim	86	13	”	"	''
kim	86	14	and	and	CC
kim	86	15	“	"	``
kim	86	16	save	save	VB
kim	86	17	lives	life	NNS
kim	86	18	,	,	,
kim	86	19	”	"	''
kim	86	20	which	which	WDT
kim	86	21	one	one	PRP
kim	86	22	should	should	MD
kim	86	23	trump	trump	VB
kim	86	24	the	the	DT
kim	86	25	other	other	JJ
kim	86	26	?	?	.
kim	87	1	The	the	DT
kim	87	2	conflict	conflict	NN
kim	87	3	among	among	IN
kim	87	4	values	value	NNS
kim	87	5	is	be	VBZ
kim	87	6	common	common	JJ
kim	87	7	in	in	IN
kim	87	8	life	life	NN
kim	87	9	,	,	,
kim	87	10	and	and	CC
kim	87	11	deontology	deontology	NN
kim	87	12	faces	face	VBZ
kim	87	13	difficulty	difficulty	NN
kim	87	14	in	in	IN
kim	87	15	guiding	guide	VBG
kim	87	16	how	how	WRB
kim	87	17	an	an	DT
kim	87	18	intelligent	intelligent	JJ
kim	87	19	agent	agent	NN
kim	87	20	is	be	VBZ
kim	87	21	to	to	TO
kim	87	22	act	act	VB
kim	87	23	in	in	IN
kim	87	24	a	a	DT
kim	87	25	tricky	tricky	JJ
kim	87	26	situation	situation	NN
kim	87	27	like	like	IN
kim	87	28	the	the	DT
kim	87	29	trolley	trolley	NN
kim	87	30	problem	problem	NN
kim	87	31	.	.	.
kim	88	1	[	[	-LRB-
kim	88	2	footnoteRef:9	footnoteRef:9	NNP
kim	88	3	]	]	-RRB-
kim	88	4	[	[	-LRB-
kim	88	5	9	9	CD
kim	88	6	:	:	:
kim	88	7	Some	some	DT
kim	88	8	moral	moral	JJ
kim	88	9	philosophers	philosopher	NNS
kim	88	10	doubt	doubt	VBP
kim	88	11	the	the	DT
kim	88	12	value	value	NN
kim	88	13	of	of	IN
kim	88	14	our	-PRON-	PRP$
kim	88	15	moral	moral	JJ
kim	88	16	intuition	intuition	NN
kim	88	17	in	in	IN
kim	88	18	constructing	construct	VBG
kim	88	19	a	a	DT
kim	88	20	moral	moral	JJ
kim	88	21	theory	theory	NN
kim	88	22	.	.	.
kim	89	1	See	see	VB
kim	89	2	Singer	singer	NN
kim	89	3	2005	2005	CD
kim	89	4	,	,	,
kim	89	5	for	for	IN
kim	89	6	example	example	NN
kim	89	7	.	.	.
kim	90	1	But	but	CC
kim	90	2	a	a	DT
kim	90	3	moral	moral	JJ
kim	90	4	theory	theory	NN
kim	90	5	that	that	IN
kim	90	6	clashes	clash	NNS
kim	90	7	with	with	IN
kim	90	8	common	common	JJ
kim	90	9	moral	moral	JJ
kim	90	10	intuition	intuition	NN
kim	90	11	is	be	VBZ
kim	90	12	unlikely	unlikely	JJ
kim	90	13	to	to	TO
kim	90	14	be	be	VB
kim	90	15	sought	seek	VBN
kim	90	16	out	out	RP
kim	90	17	as	as	IN
kim	90	18	a	a	DT
kim	90	19	guide	guide	NN
kim	90	20	to	to	IN
kim	90	21	making	make	VBG
kim	90	22	an	an	DT
kim	90	23	ethical	ethical	JJ
kim	90	24	decision	decision	NN
kim	90	25	.	.	.
kim	90	26	]	]	-RRB-
kim	91	1	3	3	LS
kim	91	2	.	.	.
kim	92	1	Understanding	understand	VBG
kim	92	2	What	what	WP
kim	92	3	Ethics	Ethics	NNP
kim	92	4	Has	have	VBZ
kim	92	5	to	to	TO
kim	92	6	Offer	offer	VB
kim	92	7	Due	due	IN
kim	92	8	to	to	IN
kim	92	9	the	the	DT
kim	92	10	high	high	JJ
kim	92	11	stakes	stake	NNS
kim	92	12	involved	involve	VBN
kim	92	13	,	,	,
kim	92	14	AI	AI	NNP
kim	92	15	-	-	HYPH
kim	92	16	powered	power	VBN
kim	92	17	military	military	JJ
kim	92	18	robots	robot	NNS
kim	92	19	and	and	CC
kim	92	20	autonomous	autonomous	JJ
kim	92	21	weapons	weapon	NNS
kim	92	22	systems	system	NNS
kim	92	23	can	can	MD
kim	92	24	present	present	VB
kim	92	25	the	the	DT
kim	92	26	moral	moral	JJ
kim	92	27	dilemma	dilemma	NN
kim	92	28	in	in	IN
kim	92	29	the	the	DT
kim	92	30	trolley	trolley	NN
kim	92	31	problem	problem	NN
kim	92	32	more	more	RBR
kim	92	33	convincingly	convincingly	RB
kim	92	34	.	.	.
kim	93	1	Suppose	suppose	VB
kim	93	2	that	that	IN
kim	93	3	some	some	DT
kim	93	4	engineers	engineer	NNS
kim	93	5	,	,	,
kim	93	6	following	follow	VBG
kim	93	7	utilitarianism	utilitarianism	NN
kim	93	8	and	and	CC
kim	93	9	interpreting	interpret	VBG
kim	93	10	victory	victory	NN
kim	93	11	as	as	IN
kim	93	12	the	the	DT
kim	93	13	ultimate	ultimate	JJ
kim	93	14	good	good	JJ
kim	93	15	/	/	SYM
kim	93	16	utility	utility	NN
kim	93	17	,	,	,
kim	93	18	wish	wish	VBP
kim	93	19	to	to	TO
kim	93	20	program	program	VB
kim	93	21	an	an	DT
kim	93	22	unmanned	unmanned	JJ
kim	93	23	aerial	aerial	JJ
kim	93	24	vehicle	vehicle	NN
kim	93	25	(	(	-LRB-
kim	93	26	UAV	UAV	NNP
kim	93	27	)	)	-RRB-
kim	93	28	to	to	TO
kim	93	29	autonomously	autonomously	RB
kim	93	30	drop	drop	VB
kim	93	31	bombs	bomb	NNS
kim	93	32	in	in	IN
kim	93	33	order	order	NN
kim	93	34	to	to	TO
kim	93	35	maximize	maximize	VB
kim	93	36	the	the	DT
kim	93	37	chances	chance	NNS
kim	93	38	of	of	IN
kim	93	39	victory	victory	NN
kim	93	40	.	.	.
kim	94	1	That	that	DT
kim	94	2	may	may	MD
kim	94	3	result	result	VB
kim	94	4	in	in	IN
kim	94	5	sacrificing	sacrifice	VBG
kim	94	6	a	a	DT
kim	94	7	greater	great	JJR
kim	94	8	number	number	NN
kim	94	9	of	of	IN
kim	94	10	civilians	civilian	NNS
kim	94	11	than	than	IN
kim	94	12	necessary	necessary	JJ
kim	94	13	,	,	,
kim	94	14	and	and	CC
kim	94	15	many	many	JJ
kim	94	16	will	will	MD
kim	94	17	consider	consider	VB
kim	94	18	this	this	DT
kim	94	19	to	to	TO
kim	94	20	be	be	VB
kim	94	21	morally	morally	RB
kim	94	22	wrong	wrong	JJ
kim	94	23	.	.	.
kim	95	1	Now	now	RB
kim	95	2	imagine	imagine	VB
kim	95	3	different	different	JJ
kim	95	4	engineers	engineer	NNS
kim	95	5	who	who	WP
kim	95	6	,	,	,
kim	95	7	adopting	adopt	VBG
kim	95	8	deontology	deontology	NN
kim	95	9	and	and	CC
kim	95	10	following	follow	VBG
kim	95	11	the	the	DT
kim	95	12	moral	moral	JJ
kim	95	13	principle	principle	NN
kim	95	14	of	of	IN
kim	95	15	not	not	RB
kim	95	16	killing	kill	VBG
kim	95	17	people	people	NNS
kim	95	18	,	,	,
kim	95	19	program	program	VB
kim	95	20	a	a	DT
kim	95	21	UAV	UAV	NNP
kim	95	22	to	to	TO
kim	95	23	autonomously	autonomously	RB
kim	95	24	act	act	VB
kim	95	25	in	in	IN
kim	95	26	a	a	DT
kim	95	27	manner	manner	NN
kim	95	28	that	that	WDT
kim	95	29	minimizes	minimize	VBZ
kim	95	30	casualties	casualty	NNS
kim	95	31	.	.	.
kim	96	1	This	this	DT
kim	96	2	may	may	MD
kim	96	3	lead	lead	VB
kim	96	4	to	to	IN
kim	96	5	defeat	defeat	NN
kim	96	6	on	on	IN
kim	96	7	the	the	DT
kim	96	8	battlefield	battlefield	NN
kim	96	9	,	,	,
kim	96	10	because	because	IN
kim	96	11	minimizing	minimize	VBG
kim	96	12	casualties	casualty	NNS
kim	96	13	may	may	MD
kim	96	14	not	not	RB
kim	96	15	be	be	VB
kim	96	16	always	always	RB
kim	96	17	advantageous	advantageous	JJ
kim	96	18	to	to	IN
kim	96	19	winning	win	VBG
kim	96	20	a	a	DT
kim	96	21	war	war	NN
kim	96	22	.	.	.
kim	97	1	These	these	DT
kim	97	2	considerations	consideration	NNS
kim	97	3	suggest	suggest	VBP
kim	97	4	that	that	IN
kim	97	5	philosophical	philosophical	JJ
kim	97	6	insights	insight	NNS
kim	97	7	from	from	IN
kim	97	8	utilitarianism	utilitarianism	NN
kim	97	9	and	and	CC
kim	97	10	deontology	deontology	NN
kim	97	11	may	may	MD
kim	97	12	provide	provide	VB
kim	97	13	little	little	JJ
kim	97	14	practical	practical	JJ
kim	97	15	guidance	guidance	NN
kim	97	16	on	on	IN
kim	97	17	how	how	WRB
kim	97	18	to	to	TO
kim	97	19	program	program	VB
kim	97	20	autonomous	autonomous	JJ
kim	97	21	AI	AI	NNP
kim	97	22	systems	system	NNS
kim	97	23	to	to	TO
kim	97	24	act	act	VB
kim	97	25	morally	morally	RB
kim	97	26	.	.	.
kim	98	1	Ethicists	ethicist	NNS
kim	98	2	seek	seek	VBP
kim	98	3	abstract	abstract	JJ
kim	98	4	principles	principle	NNS
kim	98	5	that	that	WDT
kim	98	6	can	can	MD
kim	98	7	be	be	VB
kim	98	8	generalized	generalize	VBN
kim	98	9	.	.	.
kim	99	1	For	for	IN
kim	99	2	this	this	DT
kim	99	3	reason	reason	NN
kim	99	4	,	,	,
kim	99	5	they	-PRON-	PRP
kim	99	6	are	be	VBP
kim	99	7	interested	interested	JJ
kim	99	8	in	in	IN
kim	99	9	borderline	borderline	NN
kim	99	10	cases	case	NNS
kim	99	11	that	that	WDT
kim	99	12	reveal	reveal	VBP
kim	99	13	subtle	subtle	JJ
kim	99	14	differences	difference	NNS
kim	99	15	in	in	IN
kim	99	16	our	-PRON-	PRP$
kim	99	17	moral	moral	JJ
kim	99	18	intuition	intuition	NN
kim	99	19	and	and	CC
kim	99	20	varying	vary	VBG
kim	99	21	moral	moral	JJ
kim	99	22	theories	theory	NNS
kim	99	23	.	.	.
kim	100	1	Their	-PRON-	PRP$
kim	100	2	goal	goal	NN
kim	100	3	is	be	VBZ
kim	100	4	to	to	TO
kim	100	5	define	define	VB
kim	100	6	what	what	WP
kim	100	7	is	be	VBZ
kim	100	8	moral	moral	JJ
kim	100	9	and	and	CC
kim	100	10	investigate	investigate	VB
kim	100	11	how	how	WRB
kim	100	12	moral	moral	JJ
kim	100	13	reasoning	reasoning	NN
kim	100	14	works	work	NNS
kim	100	15	or	or	CC
kim	100	16	should	should	MD
kim	100	17	work	work	VB
kim	100	18	.	.	.
kim	101	1	By	by	IN
kim	101	2	contrast	contrast	NN
kim	101	3	,	,	,
kim	101	4	engineers	engineer	NNS
kim	101	5	and	and	CC
kim	101	6	programmers	programmer	NNS
kim	101	7	pursue	pursue	VBP
kim	101	8	practical	practical	JJ
kim	101	9	solutions	solution	NNS
kim	101	10	to	to	IN
kim	101	11	real	real	JJ
kim	101	12	-	-	HYPH
kim	101	13	life	life	NN
kim	101	14	problems	problem	NNS
kim	101	15	and	and	CC
kim	101	16	look	look	VB
kim	101	17	for	for	IN
kim	101	18	guidelines	guideline	NNS
kim	101	19	that	that	WDT
kim	101	20	will	will	MD
kim	101	21	help	help	VB
kim	101	22	with	with	IN
kim	101	23	implementing	implement	VBG
kim	101	24	those	those	DT
kim	101	25	solutions	solution	NNS
kim	101	26	.	.	.
kim	102	1	Their	-PRON-	PRP$
kim	102	2	focus	focus	NN
kim	102	3	is	be	VBZ
kim	102	4	on	on	IN
kim	102	5	creating	create	VBG
kim	102	6	a	a	DT
kim	102	7	set	set	NN
kim	102	8	of	of	IN
kim	102	9	constraints	constraint	NNS
kim	102	10	and	and	CC
kim	102	11	if	if	IN
kim	102	12	-	-	HYPH
kim	102	13	then	then	RB
kim	102	14	statements	statement	NNS
kim	102	15	,	,	,
kim	102	16	which	which	WDT
kim	102	17	will	will	MD
kim	102	18	allow	allow	VB
kim	102	19	a	a	DT
kim	102	20	machine	machine	NN
kim	102	21	to	to	TO
kim	102	22	identify	identify	VB
kim	102	23	and	and	CC
kim	102	24	process	process	VB
kim	102	25	morally	morally	RB
kim	102	26	relevant	relevant	JJ
kim	102	27	considerations	consideration	NNS
kim	102	28	,	,	,
kim	102	29	so	so	IN
kim	102	30	that	that	IN
kim	102	31	it	-PRON-	PRP
kim	102	32	can	can	MD
kim	102	33	determine	determine	VB
kim	102	34	and	and	CC
kim	102	35	execute	execute	VB
kim	102	36	an	an	DT
kim	102	37	action	action	NN
kim	102	38	that	that	WDT
kim	102	39	is	be	VBZ
kim	102	40	not	not	RB
kim	102	41	only	only	RB
kim	102	42	rational	rational	JJ
kim	102	43	but	but	CC
kim	102	44	also	also	RB
kim	102	45	ethical	ethical	JJ
kim	102	46	in	in	IN
kim	102	47	the	the	DT
kim	102	48	given	give	VBN
kim	102	49	situation	situation	NN
kim	102	50	.	.	.
kim	103	1	[	[	-LRB-
kim	103	2	footnoteRef:10	footnoteref:10	CD
kim	103	3	]	]	-RRB-
kim	103	4	[	[	-LRB-
kim	103	5	10	10	CD
kim	103	6	:	:	:
kim	103	7	Note	note	VB
kim	103	8	that	that	IN
kim	103	9	this	this	DT
kim	103	10	moral	moral	JJ
kim	103	11	decision	decision	NN
kim	103	12	-	-	HYPH
kim	103	13	making	make	VBG
kim	103	14	process	process	NN
kim	103	15	can	can	MD
kim	103	16	be	be	VB
kim	103	17	modeled	model	VBN
kim	103	18	as	as	IN
kim	103	19	rule	rule	NN
kim	103	20	-	-	HYPH
kim	103	21	based	base	VBN
kim	103	22	,	,	,
kim	103	23	as	as	IN
kim	103	24	in	in	IN
kim	103	25	a	a	DT
kim	103	26	symbolic	symbolic	JJ
kim	103	27	AI	ai	NN
kim	103	28	system	system	NN
kim	103	29	,	,	,
kim	103	30	although	although	IN
kim	103	31	it	-PRON-	PRP
kim	103	32	does	do	VBZ
kim	103	33	not	not	RB
kim	103	34	have	have	VB
kim	103	35	to	to	TO
kim	103	36	be	be	VB
kim	103	37	.	.	.
kim	104	1	It	-PRON-	PRP
kim	104	2	can	can	MD
kim	104	3	be	be	VB
kim	104	4	based	base	VBN
kim	104	5	upon	upon	IN
kim	104	6	a	a	DT
kim	104	7	machine	machine	NN
kim	104	8	learning	learning	NN
kim	104	9	approach	approach	NN
kim	104	10	.	.	.
kim	105	1	It	-PRON-	PRP
kim	105	2	is	be	VBZ
kim	105	3	also	also	RB
kim	105	4	possible	possible	JJ
kim	105	5	to	to	TO
kim	105	6	combine	combine	VB
kim	105	7	both	both	DT
kim	105	8	.	.	.
kim	106	1	See	see	VB
kim	106	2	Vincent	Vincent	NNP
kim	106	3	Conitzer	Conitzer	NNP
kim	106	4	et	et	NNP
kim	106	5	al	al	NNP
kim	106	6	.	.	.
kim	107	1	2017	2017	CD
kim	107	2	.	.	.
kim	107	3	]	]	-RRB-
kim	108	1	On	on	IN
kim	108	2	the	the	DT
kim	108	3	other	other	JJ
kim	108	4	hand	hand	NN
kim	108	5	,	,	,
kim	108	6	military	military	JJ
kim	108	7	commanders	commander	NNS
kim	108	8	and	and	CC
kim	108	9	soldiers	soldier	NNS
kim	108	10	might	may	MD
kim	108	11	aim	aim	VB
kim	108	12	to	to	TO
kim	108	13	end	end	VB
kim	108	14	a	a	DT
kim	108	15	conflict	conflict	NN
kim	108	16	,	,	,
kim	108	17	bring	bring	VB
kim	108	18	peace	peace	NN
kim	108	19	,	,	,
kim	108	20	and	and	CC
kim	108	21	facilitate	facilitate	VB
kim	108	22	restoring	restore	VBG
kim	108	23	and	and	CC
kim	108	24	establishing	establish	VBG
kim	108	25	universally	universally	RB
kim	108	26	recognized	recognize	VBN
kim	108	27	human	human	JJ
kim	108	28	values	value	NNS
kim	108	29	such	such	JJ
kim	108	30	as	as	IN
kim	108	31	freedom	freedom	NN
kim	108	32	,	,	,
kim	108	33	equality	equality	NN
kim	108	34	,	,	,
kim	108	35	justice	justice	NN
kim	108	36	,	,	,
kim	108	37	and	and	CC
kim	108	38	self	self	NN
kim	108	39	-	-	HYPH
kim	108	40	determination	determination	NN
kim	108	41	.	.	.
kim	109	1	In	in	IN
kim	109	2	order	order	NN
kim	109	3	to	to	TO
kim	109	4	achieve	achieve	VB
kim	109	5	this	this	DT
kim	109	6	goal	goal	NN
kim	109	7	,	,	,
kim	109	8	they	-PRON-	PRP
kim	109	9	must	must	MD
kim	109	10	make	make	VB
kim	109	11	the	the	DT
kim	109	12	best	good	JJS
kim	109	13	strategic	strategic	JJ
kim	109	14	decisions	decision	NNS
kim	109	15	and	and	CC
kim	109	16	take	take	VB
kim	109	17	the	the	DT
kim	109	18	most	most	RBS
kim	109	19	appropriate	appropriate	JJ
kim	109	20	actions	action	NNS
kim	109	21	.	.	.
kim	110	1	In	in	IN
kim	110	2	deciding	decide	VBG
kim	110	3	on	on	IN
kim	110	4	those	those	DT
kim	110	5	actions	action	NNS
kim	110	6	,	,	,
kim	110	7	they	-PRON-	PRP
kim	110	8	are	be	VBP
kim	110	9	also	also	RB
kim	110	10	responsible	responsible	JJ
kim	110	11	for	for	IN
kim	110	12	abiding	abide	VBG
kim	110	13	by	by	IN
kim	110	14	the	the	DT
kim	110	15	principles	principle	NNS
kim	110	16	of	of	IN
kim	110	17	jus	jus	NN
kim	110	18	in	in	IN
kim	110	19	bello	bello	NN
kim	110	20	and	and	CC
kim	110	21	for	for	IN
kim	110	22	not	not	RB
kim	110	23	abdicating	abdicate	VBG
kim	110	24	their	-PRON-	PRP$
kim	110	25	moral	moral	JJ
kim	110	26	responsibility	responsibility	NN
kim	110	27	,	,	,
kim	110	28	protecting	protect	VBG
kim	110	29	civilians	civilian	NNS
kim	110	30	and	and	CC
kim	110	31	minimizing	minimize	VBG
kim	110	32	harm	harm	NN
kim	110	33	,	,	,
kim	110	34	violence	violence	NN
kim	110	35	,	,	,
kim	110	36	and	and	CC
kim	110	37	destruction	destruction	NN
kim	110	38	as	as	RB
kim	110	39	much	much	RB
kim	110	40	as	as	IN
kim	110	41	possible	possible	JJ
kim	110	42	.	.	.
kim	111	1	[	[	-LRB-
kim	111	2	footnoteRef:11	footnoteRef:11	NNP
kim	111	3	]	]	-RRB-
kim	111	4	The	the	DT
kim	111	5	goal	goal	NN
kim	111	6	of	of	IN
kim	111	7	military	military	JJ
kim	111	8	commanders	commander	NNS
kim	111	9	and	and	CC
kim	111	10	soldiers	soldier	NNS
kim	111	11	,	,	,
kim	111	12	therefore	therefore	RB
kim	111	13	,	,	,
kim	111	14	differs	differ	NNS
kim	111	15	from	from	IN
kim	111	16	those	those	DT
kim	111	17	of	of	IN
kim	111	18	moral	moral	JJ
kim	111	19	philosophers	philosopher	NNS
kim	111	20	or	or	CC
kim	111	21	of	of	IN
kim	111	22	the	the	DT
kim	111	23	engineers	engineer	NNS
kim	111	24	of	of	IN
kim	111	25	autonomous	autonomous	JJ
kim	111	26	weapons	weapon	NNS
kim	111	27	.	.	.
kim	112	1	They	-PRON-	PRP
kim	112	2	are	be	VBP
kim	112	3	obligated	obligate	VBN
kim	112	4	to	to	TO
kim	112	5	make	make	VB
kim	112	6	quick	quick	JJ
kim	112	7	decisions	decision	NNS
kim	112	8	in	in	IN
kim	112	9	a	a	DT
kim	112	10	life	life	NN
kim	112	11	-	-	HYPH
kim	112	12	or	or	CC
kim	112	13	-	-	HYPH
kim	112	14	death	death	NN
kim	112	15	situation	situation	NN
kim	112	16	while	while	IN
kim	112	17	working	work	VBG
kim	112	18	with	with	IN
kim	112	19	AI	AI	NNP
kim	112	20	-	-	HYPH
kim	112	21	powered	power	VBN
kim	112	22	military	military	JJ
kim	112	23	systems	system	NNS
kim	112	24	.	.	.
kim	113	1	[	[	-LRB-
kim	113	2	11	11	CD
kim	113	3	:	:	:
kim	113	4	For	for	IN
kim	113	5	the	the	DT
kim	113	6	principles	principle	NNS
kim	113	7	of	of	IN
kim	113	8	jus	jus	NN
kim	113	9	in	in	IN
kim	113	10	bello	bello	NN
kim	113	11	,	,	,
kim	113	12	see	see	VB
kim	113	13	International	International	NNP
kim	113	14	Committee	Committee	NNP
kim	113	15	of	of	IN
kim	113	16	the	the	DT
kim	113	17	Red	Red	NNP
kim	113	18	Cross	Cross	NNP
kim	113	19	2015	2015	CD
kim	113	20	.	.	.
kim	113	21	]	]	-RRB-
kim	114	1	These	these	DT
kim	114	2	different	different	JJ
kim	114	3	goals	goal	NNS
kim	114	4	and	and	CC
kim	114	5	interests	interest	NNS
kim	114	6	explain	explain	VBP
kim	114	7	why	why	WRB
kim	114	8	moral	moral	JJ
kim	114	9	philosophers	philosopher	NNS
kim	114	10	’	’	POS
kim	114	11	discussion	discussion	NN
kim	114	12	on	on	IN
kim	114	13	the	the	DT
kim	114	14	trolley	trolley	NN
kim	114	15	problem	problem	NN
kim	114	16	may	may	MD
kim	114	17	be	be	VB
kim	114	18	disappointing	disappointing	JJ
kim	114	19	to	to	IN
kim	114	20	AI	AI	NNP
kim	114	21	programmers	programmer	NNS
kim	114	22	or	or	CC
kim	114	23	military	military	JJ
kim	114	24	commanders	commander	NNS
kim	114	25	and	and	CC
kim	114	26	soldiers	soldier	NNS
kim	114	27	.	.	.
kim	115	1	Ethics	ethic	NNS
kim	115	2	does	do	VBZ
kim	115	3	not	not	RB
kim	115	4	provide	provide	VB
kim	115	5	an	an	DT
kim	115	6	easy	easy	JJ
kim	115	7	answer	answer	NN
kim	115	8	to	to	IN
kim	115	9	the	the	DT
kim	115	10	question	question	NN
kim	115	11	of	of	IN
kim	115	12	how	how	WRB
kim	115	13	one	one	PRP
kim	115	14	should	should	MD
kim	115	15	program	program	VB
kim	115	16	moral	moral	JJ
kim	115	17	decision	decision	NN
kim	115	18	-	-	HYPH
kim	115	19	making	making	NN
kim	115	20	into	into	IN
kim	115	21	intelligent	intelligent	JJ
kim	115	22	machines	machine	NNS
kim	115	23	.	.	.
kim	116	1	Nor	nor	CC
kim	116	2	does	do	VBZ
kim	116	3	it	-PRON-	PRP
kim	116	4	prescribe	prescribe	VB
kim	116	5	the	the	DT
kim	116	6	right	right	JJ
kim	116	7	moral	moral	JJ
kim	116	8	decision	decision	NN
kim	116	9	in	in	IN
kim	116	10	a	a	DT
kim	116	11	battlefield	battlefield	NN
kim	116	12	.	.	.
kim	117	1	But	but	CC
kim	117	2	taking	take	VBG
kim	117	3	this	this	DT
kim	117	4	as	as	IN
kim	117	5	a	a	DT
kim	117	6	shortcoming	shortcoming	NN
kim	117	7	of	of	IN
kim	117	8	ethics	ethic	NNS
kim	117	9	is	be	VBZ
kim	117	10	missing	miss	VBG
kim	117	11	the	the	DT
kim	117	12	point	point	NN
kim	117	13	.	.	.
kim	118	1	The	the	DT
kim	118	2	role	role	NN
kim	118	3	of	of	IN
kim	118	4	moral	moral	JJ
kim	118	5	philosophy	philosophy	NN
kim	118	6	is	be	VBZ
kim	118	7	not	not	RB
kim	118	8	to	to	TO
kim	118	9	make	make	VB
kim	118	10	decision	decision	NN
kim	118	11	-	-	HYPH
kim	118	12	making	make	VBG
kim	118	13	easier	easy	JJR
kim	118	14	but	but	CC
kim	118	15	to	to	TO
kim	118	16	highlight	highlight	VB
kim	118	17	and	and	CC
kim	118	18	articulate	articulate	VB
kim	118	19	the	the	DT
kim	118	20	difficulty	difficulty	NN
kim	118	21	and	and	CC
kim	118	22	complexity	complexity	NN
kim	118	23	involved	involve	VBN
kim	118	24	in	in	IN
kim	118	25	it	-PRON-	PRP
kim	118	26	.	.	.
kim	119	1	4	4	LS
kim	119	2	.	.	.
kim	120	1	Ethical	ethical	JJ
kim	120	2	Challenges	Challenges	NNPS
kim	120	3	from	from	IN
kim	120	4	Autonomous	Autonomous	NNP
kim	120	5	AI	AI	NNP
kim	120	6	Systems	Systems	NNP
kim	120	7	The	the	DT
kim	120	8	complexity	complexity	NN
kim	120	9	of	of	IN
kim	120	10	ethical	ethical	JJ
kim	120	11	questions	question	NNS
kim	120	12	means	mean	VBZ
kim	120	13	that	that	IN
kim	120	14	dealing	deal	VBG
kim	120	15	with	with	IN
kim	120	16	the	the	DT
kim	120	17	morality	morality	NN
kim	120	18	of	of	IN
kim	120	19	an	an	DT
kim	120	20	action	action	NN
kim	120	21	by	by	IN
kim	120	22	an	an	DT
kim	120	23	autonomous	autonomous	JJ
kim	120	24	AI	ai	NN
kim	120	25	system	system	NN
kim	120	26	will	will	MD
kim	120	27	require	require	VB
kim	120	28	more	more	JJR
kim	120	29	than	than	IN
kim	120	30	a	a	DT
kim	120	31	clever	clever	JJ
kim	120	32	engineering	engineering	NN
kim	120	33	or	or	CC
kim	120	34	programming	programming	NN
kim	120	35	solution	solution	NN
kim	120	36	.	.	.
kim	121	1	The	the	DT
kim	121	2	fact	fact	NN
kim	121	3	that	that	IN
kim	121	4	ethics	ethic	NNS
kim	121	5	does	do	VBZ
kim	121	6	not	not	RB
kim	121	7	eliminate	eliminate	VB
kim	121	8	the	the	DT
kim	121	9	inherent	inherent	JJ
kim	121	10	ambiguity	ambiguity	NN
kim	121	11	in	in	IN
kim	121	12	many	many	JJ
kim	121	13	moral	moral	JJ
kim	121	14	decisions	decision	NNS
kim	121	15	should	should	MD
kim	121	16	not	not	RB
kim	121	17	lead	lead	VB
kim	121	18	to	to	IN
kim	121	19	the	the	DT
kim	121	20	dismissal	dismissal	NN
kim	121	21	of	of	IN
kim	121	22	ethical	ethical	JJ
kim	121	23	challenges	challenge	NNS
kim	121	24	from	from	IN
kim	121	25	autonomous	autonomous	JJ
kim	121	26	AI	AI	NNP
kim	121	27	systems	system	NNS
kim	121	28	.	.	.
kim	122	1	By	by	IN
kim	122	2	injecting	inject	VBG
kim	122	3	the	the	DT
kim	122	4	capacity	capacity	NN
kim	122	5	for	for	IN
kim	122	6	autonomous	autonomous	JJ
kim	122	7	decision	decision	NN
kim	122	8	-	-	HYPH
kim	122	9	making	making	NN
kim	122	10	into	into	IN
kim	122	11	major	major	JJ
kim	122	12	systems	system	NNS
kim	122	13	and	and	CC
kim	122	14	tools	tool	NNS
kim	122	15	,	,	,
kim	122	16	AI	AI	NNP
kim	122	17	can	can	MD
kim	122	18	fundamentally	fundamentally	RB
kim	122	19	transform	transform	VB
kim	122	20	any	any	DT
kim	122	21	given	give	VBN
kim	122	22	field	field	NN
kim	122	23	.	.	.
kim	123	1	For	for	IN
kim	123	2	example	example	NN
kim	123	3	,	,	,
kim	123	4	AI	AI	NNP
kim	123	5	-	-	HYPH
kim	123	6	powered	power	VBN
kim	123	7	military	military	JJ
kim	123	8	robots	robot	NNS
kim	123	9	are	be	VBP
kim	123	10	not	not	RB
kim	123	11	just	just	RB
kim	123	12	another	another	DT
kim	123	13	kind	kind	NN
kim	123	14	of	of	IN
kim	123	15	weapon	weapon	NN
kim	123	16	.	.	.
kim	124	1	When	when	WRB
kim	124	2	widely	widely	RB
kim	124	3	deployed	deploy	VBN
kim	124	4	,	,	,
kim	124	5	they	-PRON-	PRP
kim	124	6	can	can	MD
kim	124	7	change	change	VB
kim	124	8	the	the	DT
kim	124	9	nature	nature	NN
kim	124	10	of	of	IN
kim	124	11	war	war	NN
kim	124	12	itself	-PRON-	PRP
kim	124	13	.	.	.
kim	125	1	Described	describe	VBN
kim	125	2	below	below	IN
kim	125	3	are	be	VBP
kim	125	4	some	some	DT
kim	125	5	of	of	IN
kim	125	6	the	the	DT
kim	125	7	significant	significant	JJ
kim	125	8	ethical	ethical	JJ
kim	125	9	challenges	challenge	NNS
kim	125	10	that	that	WDT
kim	125	11	autonomous	autonomous	JJ
kim	125	12	AI	AI	NNP
kim	125	13	systems	system	NNS
kim	125	14	such	such	JJ
kim	125	15	as	as	IN
kim	125	16	military	military	JJ
kim	125	17	robots	robot	NNS
kim	125	18	present	present	JJ
kim	125	19	.	.	.
kim	126	1	(	(	-LRB-
kim	126	2	a	a	LS
kim	126	3	)	)	-RRB-
kim	126	4	Moral	moral	JJ
kim	126	5	desensitization	desensitization	NN
kim	126	6	AI	ai	NN
kim	126	7	-	-	HYPH
kim	126	8	powered	power	VBN
kim	126	9	military	military	JJ
kim	126	10	robots	robot	NNS
kim	126	11	go	go	VBP
kim	126	12	further	further	RB
kim	126	13	than	than	IN
kim	126	14	any	any	DT
kim	126	15	remotely	remotely	RB
kim	126	16	-	-	HYPH
kim	126	17	operated	operate	VBN
kim	126	18	weapons	weapon	NNS
kim	126	19	.	.	.
kim	127	1	They	-PRON-	PRP
kim	127	2	can	can	MD
kim	127	3	identify	identify	VB
kim	127	4	a	a	DT
kim	127	5	target	target	NN
kim	127	6	and	and	CC
kim	127	7	initiate	initiate	VB
kim	127	8	an	an	DT
kim	127	9	attack	attack	NN
kim	127	10	on	on	IN
kim	127	11	their	-PRON-	PRP$
kim	127	12	own	own	JJ
kim	127	13	.	.	.
kim	128	1	Due	due	IN
kim	128	2	to	to	IN
kim	128	3	their	-PRON-	PRP$
kim	128	4	autonomy	autonomy	NN
kim	128	5	,	,	,
kim	128	6	military	military	JJ
kim	128	7	robots	robot	NNS
kim	128	8	can	can	MD
kim	128	9	significantly	significantly	RB
kim	128	10	increase	increase	VB
kim	128	11	the	the	DT
kim	128	12	distance	distance	NN
kim	128	13	between	between	IN
kim	128	14	the	the	DT
kim	128	15	party	party	NN
kim	128	16	that	that	WDT
kim	128	17	kills	kill	VBZ
kim	128	18	and	and	CC
kim	128	19	the	the	DT
kim	128	20	party	party	NN
kim	128	21	that	that	WDT
kim	128	22	gets	get	VBZ
kim	128	23	killed	kill	VBN
kim	128	24	(	(	-LRB-
kim	128	25	Sharkey	Sharkey	NNP
kim	128	26	2012	2012	CD
kim	128	27	)	)	-RRB-
kim	128	28	.	.	.
kim	129	1	This	this	DT
kim	129	2	increase	increase	NN
kim	129	3	,	,	,
kim	129	4	however	however	RB
kim	129	5	,	,	,
kim	129	6	may	may	MD
kim	129	7	lead	lead	VB
kim	129	8	people	people	NNS
kim	129	9	to	to	TO
kim	129	10	surrender	surrender	VB
kim	129	11	their	-PRON-	PRP$
kim	129	12	own	own	JJ
kim	129	13	moral	moral	JJ
kim	129	14	responsibility	responsibility	NN
kim	129	15	to	to	IN
kim	129	16	a	a	DT
kim	129	17	machine	machine	NN
kim	129	18	,	,	,
kim	129	19	thereby	thereby	RB
kim	129	20	resulting	result	VBG
kim	129	21	in	in	IN
kim	129	22	the	the	DT
kim	129	23	loss	loss	NN
kim	129	24	of	of	IN
kim	129	25	humanity	humanity	NN
kim	129	26	,	,	,
kim	129	27	which	which	WDT
kim	129	28	is	be	VBZ
kim	129	29	a	a	DT
kim	129	30	serious	serious	JJ
kim	129	31	moral	moral	JJ
kim	129	32	risk	risk	NN
kim	129	33	(	(	-LRB-
kim	129	34	Davis	Davis	NNP
kim	129	35	2007	2007	CD
kim	129	36	)	)	-RRB-
kim	129	37	.	.	.
kim	130	1	The	the	DT
kim	130	2	more	more	RBR
kim	130	3	autonomous	autonomous	JJ
kim	130	4	military	military	JJ
kim	130	5	robots	robot	NNS
kim	130	6	become	become	VBP
kim	130	7	,	,	,
kim	130	8	the	the	DT
kim	130	9	less	less	JJR
kim	130	10	responsibility	responsibility	NN
kim	130	11	for	for	IN
kim	130	12	life	life	NN
kim	130	13	-	-	HYPH
kim	130	14	or	or	CC
kim	130	15	-	-	HYPH
kim	130	16	death	death	NN
kim	130	17	decisions	decision	NNS
kim	130	18	may	may	MD
kim	130	19	come	come	VB
kim	130	20	to	to	TO
kim	130	21	rest	rest	VB
kim	130	22	with	with	IN
kim	130	23	humans	human	NNS
kim	130	24	.	.	.
kim	131	1	(	(	-LRB-
kim	131	2	b	b	NN
kim	131	3	)	)	-RRB-
kim	131	4	Unintended	Unintended	NNP
kim	131	5	outcomes	outcome	NNS
kim	131	6	As	as	IN
kim	131	7	military	military	JJ
kim	131	8	robots	robot	NNS
kim	131	9	make	make	VBP
kim	131	10	killing	kill	VBG
kim	131	11	easier	easy	JJR
kim	131	12	,	,	,
kim	131	13	small	small	JJ
kim	131	14	conflicts	conflict	NNS
kim	131	15	may	may	MD
kim	131	16	more	more	RBR
kim	131	17	quickly	quickly	RB
kim	131	18	escalate	escalate	VB
kim	131	19	to	to	IN
kim	131	20	a	a	DT
kim	131	21	war	war	NN
kim	131	22	.	.	.
kim	132	1	The	the	DT
kim	132	2	side	side	NN
kim	132	3	that	that	WDT
kim	132	4	deploys	deploy	VBZ
kim	132	5	AI	ai	NN
kim	132	6	-	-	HYPH
kim	132	7	powered	power	VBN
kim	132	8	military	military	JJ
kim	132	9	robots	robot	NNS
kim	132	10	is	be	VBZ
kim	132	11	likely	likely	JJ
kim	132	12	to	to	TO
kim	132	13	suffer	suffer	VB
kim	132	14	fewer	few	JJR
kim	132	15	casualties	casualty	NNS
kim	132	16	itself	-PRON-	PRP
kim	132	17	while	while	IN
kim	132	18	inflicting	inflict	VBG
kim	132	19	more	more	JJR
kim	132	20	casualties	casualty	NNS
kim	132	21	on	on	IN
kim	132	22	the	the	DT
kim	132	23	enemy	enemy	NN
kim	132	24	side	side	NN
kim	132	25	.	.	.
kim	133	1	This	this	DT
kim	133	2	may	may	MD
kim	133	3	make	make	VB
kim	133	4	the	the	DT
kim	133	5	military	military	NN
kim	133	6	more	more	RBR
kim	133	7	inclined	inclined	JJ
kim	133	8	to	to	TO
kim	133	9	start	start	VB
kim	133	10	a	a	DT
kim	133	11	war	war	NN
kim	133	12	.	.	.
kim	134	1	Ironically	ironically	RB
kim	134	2	,	,	,
kim	134	3	when	when	WRB
kim	134	4	everyone	everyone	NN
kim	134	5	thinks	think	VBZ
kim	134	6	and	and	CC
kim	134	7	acts	act	VBZ
kim	134	8	this	this	DT
kim	134	9	way	way	NN
kim	134	10	,	,	,
kim	134	11	the	the	DT
kim	134	12	number	number	NN
kim	134	13	of	of	IN
kim	134	14	wars	war	NNS
kim	134	15	and	and	CC
kim	134	16	the	the	DT
kim	134	17	overall	overall	JJ
kim	134	18	amount	amount	NN
kim	134	19	of	of	IN
kim	134	20	violence	violence	NN
kim	134	21	and	and	CC
kim	134	22	destruction	destruction	NN
kim	134	23	in	in	IN
kim	134	24	the	the	DT
kim	134	25	world	world	NN
kim	134	26	will	will	MD
kim	134	27	only	only	RB
kim	134	28	increase	increase	VB
kim	134	29	.	.	.
kim	135	1	[	[	-LRB-
kim	135	2	footnoteRef:12	footnoteref:12	NN
kim	135	3	]	]	-RRB-
kim	135	4	[	[	-LRB-
kim	135	5	12	12	CD
kim	135	6	:	:	:
kim	135	7	Kahn	Kahn	NNP
kim	135	8	(	(	-LRB-
kim	135	9	2012	2012	CD
kim	135	10	)	)	-RRB-
kim	135	11	also	also	RB
kim	135	12	argues	argue	VBZ
kim	135	13	that	that	IN
kim	135	14	the	the	DT
kim	135	15	resulting	result	VBG
kim	135	16	increase	increase	NN
kim	135	17	in	in	IN
kim	135	18	the	the	DT
kim	135	19	number	number	NN
kim	135	20	of	of	IN
kim	135	21	wars	war	NNS
kim	135	22	by	by	IN
kim	135	23	the	the	DT
kim	135	24	use	use	NN
kim	135	25	of	of	IN
kim	135	26	military	military	JJ
kim	135	27	robots	robot	NNS
kim	135	28	will	will	MD
kim	135	29	be	be	VB
kim	135	30	morally	morally	RB
kim	135	31	bad	bad	JJ
kim	135	32	.	.	.
kim	135	33	]	]	-RRB-
kim	136	1	(	(	-LRB-
kim	136	2	c	c	NN
kim	136	3	)	)	-RRB-
kim	136	4	Surrender	surrender	NN
kim	136	5	of	of	IN
kim	136	6	moral	moral	JJ
kim	136	7	agency	agency	NN
kim	136	8	Military	military	JJ
kim	136	9	robots	robot	NNS
kim	136	10	may	may	MD
kim	136	11	fail	fail	VB
kim	136	12	to	to	TO
kim	136	13	distinguish	distinguish	VB
kim	136	14	innocents	innocent	NNS
kim	136	15	from	from	IN
kim	136	16	combatants	combatant	NNS
kim	136	17	and	and	CC
kim	136	18	kill	kill	VB
kim	136	19	the	the	DT
kim	136	20	former	former	JJ
kim	136	21	.	.	.
kim	137	1	In	in	IN
kim	137	2	such	such	PDT
kim	137	3	a	a	DT
kim	137	4	case	case	NN
kim	137	5	,	,	,
kim	137	6	can	can	MD
kim	137	7	we	-PRON-	PRP
kim	137	8	be	be	VB
kim	137	9	justified	justify	VBN
kim	137	10	in	in	IN
kim	137	11	letting	let	VBG
kim	137	12	robots	robot	NNS
kim	137	13	take	take	VB
kim	137	14	the	the	DT
kim	137	15	lives	life	NNS
kim	137	16	of	of	IN
kim	137	17	other	other	JJ
kim	137	18	human	human	JJ
kim	137	19	beings	being	NNS
kim	137	20	?	?	.
kim	138	1	Some	some	DT
kim	138	2	may	may	MD
kim	138	3	argue	argue	VB
kim	138	4	that	that	IN
kim	138	5	only	only	RB
kim	138	6	humans	human	NNS
kim	138	7	should	should	MD
kim	138	8	decide	decide	VB
kim	138	9	to	to	TO
kim	138	10	kill	kill	VB
kim	138	11	other	other	JJ
kim	138	12	humans	human	NNS
kim	138	13	,	,	,
kim	138	14	not	not	RB
kim	138	15	machines	machine	NNS
kim	138	16	(	(	-LRB-
kim	138	17	Davis	Davis	NNP
kim	138	18	2007	2007	CD
kim	138	19	)	)	-RRB-
kim	138	20	.	.	.
kim	139	1	Is	be	VBZ
kim	139	2	it	-PRON-	PRP
kim	139	3	permissible	permissible	JJ
kim	139	4	for	for	IN
kim	139	5	people	people	NNS
kim	139	6	to	to	TO
kim	139	7	delegate	delegate	VB
kim	139	8	such	such	PDT
kim	139	9	a	a	DT
kim	139	10	decision	decision	NN
kim	139	11	to	to	IN
kim	139	12	robots	robot	NNS
kim	139	13	?	?	.
kim	140	1	(	(	-LRB-
kim	140	2	d	d	LS
kim	140	3	)	)	-RRB-
kim	140	4	Opacity	opacity	NN
kim	140	5	in	in	IN
kim	140	6	decision	decision	NN
kim	140	7	-	-	HYPH
kim	140	8	making	make	VBG
kim	140	9	Machine	Machine	NNP
kim	140	10	learning	learning	NN
kim	140	11	is	be	VBZ
kim	140	12	used	use	VBN
kim	140	13	to	to	TO
kim	140	14	build	build	VB
kim	140	15	many	many	JJ
kim	140	16	AI	AI	NNP
kim	140	17	systems	system	NNS
kim	140	18	today	today	NN
kim	140	19	.	.	.
kim	141	1	Instead	instead	RB
kim	141	2	of	of	IN
kim	141	3	prescribing	prescribe	VBG
kim	141	4	a	a	DT
kim	141	5	pre	pre	JJ
kim	141	6	-	-	JJ
kim	141	7	determined	determined	JJ
kim	141	8	algorithm	algorithm	NN
kim	141	9	,	,	,
kim	141	10	a	a	DT
kim	141	11	machine	machine	NN
kim	141	12	learning	learning	NN
kim	141	13	system	system	NN
kim	141	14	goes	go	VBZ
kim	141	15	through	through	IN
kim	141	16	a	a	DT
kim	141	17	so	so	RB
kim	141	18	-	-	HYPH
kim	141	19	called	call	VBN
kim	141	20	‘	'	``
kim	141	21	training	training	NN
kim	141	22	’	'	''
kim	141	23	process	process	NN
kim	141	24	in	in	IN
kim	141	25	order	order	NN
kim	141	26	to	to	TO
kim	141	27	produce	produce	VB
kim	141	28	the	the	DT
kim	141	29	final	final	JJ
kim	141	30	algorithm	algorithm	NN
kim	141	31	from	from	IN
kim	141	32	a	a	DT
kim	141	33	large	large	JJ
kim	141	34	amount	amount	NN
kim	141	35	of	of	IN
kim	141	36	data	datum	NNS
kim	141	37	.	.	.
kim	142	1	For	for	IN
kim	142	2	example	example	NN
kim	142	3	,	,	,
kim	142	4	a	a	DT
kim	142	5	machine	machine	NN
kim	142	6	learning	learning	NN
kim	142	7	system	system	NN
kim	142	8	may	may	MD
kim	142	9	generate	generate	VB
kim	142	10	an	an	DT
kim	142	11	algorithm	algorithm	NN
kim	142	12	that	that	WDT
kim	142	13	successfully	successfully	RB
kim	142	14	recognizes	recognize	VBZ
kim	142	15	cats	cat	NNS
kim	142	16	in	in	IN
kim	142	17	a	a	DT
kim	142	18	photo	photo	NN
kim	142	19	after	after	IN
kim	142	20	going	go	VBG
kim	142	21	through	through	IN
kim	142	22	millions	million	NNS
kim	142	23	of	of	IN
kim	142	24	photos	photo	NNS
kim	142	25	that	that	WDT
kim	142	26	show	show	VBP
kim	142	27	cats	cat	NNS
kim	142	28	in	in	IN
kim	142	29	many	many	JJ
kim	142	30	different	different	JJ
kim	142	31	postures	posture	NNS
kim	142	32	from	from	IN
kim	142	33	various	various	JJ
kim	142	34	angles	angle	NNS
kim	142	35	.	.	.
kim	143	1	[	[	-LRB-
kim	143	2	footnoteRef:13	footnoteref:13	NN
kim	143	3	]	]	-RRB-
kim	143	4	But	but	CC
kim	143	5	the	the	DT
kim	143	6	resulting	result	VBG
kim	143	7	algorithm	algorithm	NN
kim	143	8	is	be	VBZ
kim	143	9	a	a	DT
kim	143	10	complex	complex	JJ
kim	143	11	mathematical	mathematical	JJ
kim	143	12	formula	formula	NN
kim	143	13	and	and	CC
kim	143	14	not	not	RB
kim	143	15	something	something	NN
kim	143	16	that	that	WDT
kim	143	17	humans	human	NNS
kim	143	18	can	can	MD
kim	143	19	easily	easily	RB
kim	143	20	decipher	decipher	VB
kim	143	21	.	.	.
kim	144	1	This	this	DT
kim	144	2	means	mean	VBZ
kim	144	3	that	that	IN
kim	144	4	the	the	DT
kim	144	5	inner	inner	JJ
kim	144	6	workings	working	NNS
kim	144	7	of	of	IN
kim	144	8	a	a	DT
kim	144	9	machine	machine	NN
kim	144	10	learning	learn	VBG
kim	144	11	AI	ai	NN
kim	144	12	system	system	NN
kim	144	13	and	and	CC
kim	144	14	its	-PRON-	PRP$
kim	144	15	decision	decision	NN
kim	144	16	-	-	HYPH
kim	144	17	making	make	VBG
kim	144	18	process	process	NN
kim	144	19	is	be	VBZ
kim	144	20	opaque	opaque	JJ
kim	144	21	to	to	IN
kim	144	22	human	human	JJ
kim	144	23	understanding	understanding	NN
kim	144	24	,	,	,
kim	144	25	even	even	RB
kim	144	26	to	to	IN
kim	144	27	those	those	DT
kim	144	28	who	who	WP
kim	144	29	built	build	VBD
kim	144	30	the	the	DT
kim	144	31	system	system	NN
kim	144	32	itself	-PRON-	PRP
kim	144	33	(	(	-LRB-
kim	144	34	Knight	Knight	NNP
kim	144	35	2017	2017	CD
kim	144	36	)	)	-RRB-
kim	144	37	.	.	.
kim	145	1	In	in	IN
kim	145	2	cases	case	NNS
kim	145	3	where	where	WRB
kim	145	4	the	the	DT
kim	145	5	actions	action	NNS
kim	145	6	of	of	IN
kim	145	7	an	an	DT
kim	145	8	AI	AI	NNP
kim	145	9	system	system	NN
kim	145	10	can	can	MD
kim	145	11	have	have	VB
kim	145	12	grave	grave	JJ
kim	145	13	consequences	consequence	NNS
kim	145	14	such	such	JJ
kim	145	15	as	as	IN
kim	145	16	a	a	DT
kim	145	17	military	military	JJ
kim	145	18	robot	robot	NN
kim	145	19	,	,	,
kim	145	20	such	such	JJ
kim	145	21	opacity	opacity	NN
kim	145	22	becomes	become	VBZ
kim	145	23	a	a	DT
kim	145	24	serious	serious	JJ
kim	145	25	problem	problem	NN
kim	145	26	.	.	.
kim	146	1	[	[	-LRB-
kim	146	2	footnoteRef:14	footnoteRef:14	NNP
kim	146	3	]	]	-RRB-
kim	146	4	In	in	IN
kim	146	5	spite	spite	NN
kim	146	6	of	of	IN
kim	146	7	these	these	DT
kim	146	8	ethical	ethical	JJ
kim	146	9	concerns	concern	NNS
kim	146	10	,	,	,
kim	146	11	autonomous	autonomous	JJ
kim	146	12	AI	AI	NNP
kim	146	13	systems	system	NNS
kim	146	14	are	be	VBP
kim	146	15	likely	likely	JJ
kim	146	16	to	to	TO
kim	146	17	continue	continue	VB
kim	146	18	to	to	TO
kim	146	19	be	be	VB
kim	146	20	developed	develop	VBN
kim	146	21	and	and	CC
kim	146	22	adopted	adopt	VBN
kim	146	23	in	in	IN
kim	146	24	many	many	JJ
kim	146	25	areas	area	NNS
kim	146	26	as	as	IN
kim	146	27	a	a	DT
kim	146	28	way	way	NN
kim	146	29	to	to	TO
kim	146	30	increase	increase	VB
kim	146	31	efficiency	efficiency	NN
kim	146	32	and	and	CC
kim	146	33	lower	low	JJR
kim	146	34	cost	cost	NN
kim	146	35	.	.	.
kim	147	1	[	[	-LRB-
kim	147	2	13	13	CD
kim	147	3	:	:	:
kim	147	4	Google	Google	NNP
kim	147	5	’s	’s	POS
kim	147	6	research	research	NN
kim	147	7	team	team	NN
kim	147	8	created	create	VBD
kim	147	9	an	an	DT
kim	147	10	AI	AI	NNP
kim	147	11	algorithm	algorithm	NN
kim	147	12	that	that	WDT
kim	147	13	learned	learn	VBD
kim	147	14	how	how	WRB
kim	147	15	to	to	TO
kim	147	16	recognize	recognize	VB
kim	147	17	a	a	DT
kim	147	18	cat	cat	NN
kim	147	19	in	in	IN
kim	147	20	2012	2012	CD
kim	147	21	.	.	.
kim	148	1	The	the	DT
kim	148	2	neural	neural	JJ
kim	148	3	network	network	NN
kim	148	4	behind	behind	IN
kim	148	5	this	this	DT
kim	148	6	algorithm	algorithm	NN
kim	148	7	had	have	VBD
kim	148	8	an	an	DT
kim	148	9	array	array	NN
kim	148	10	of	of	IN
kim	148	11	16,000	16,000	CD
kim	148	12	processors	processor	NNS
kim	148	13	and	and	CC
kim	148	14	more	more	JJR
kim	148	15	than	than	IN
kim	148	16	one	one	CD
kim	148	17	billion	billion	CD
kim	148	18	connections	connection	NNS
kim	148	19	,	,	,
kim	148	20	and	and	CC
kim	148	21	unlabeled	unlabeled	JJ
kim	148	22	random	random	JJ
kim	148	23	thumbnail	thumbnail	NN
kim	148	24	images	image	NNS
kim	148	25	from	from	IN
kim	148	26	10	10	CD
kim	148	27	million	million	CD
kim	148	28	YouTube	YouTube	NNP
kim	148	29	videos	video	NNS
kim	148	30	allowed	allow	VBD
kim	148	31	this	this	DT
kim	148	32	algorithm	algorithm	NN
kim	148	33	to	to	TO
kim	148	34	learn	learn	VB
kim	148	35	to	to	TO
kim	148	36	identify	identify	VB
kim	148	37	cats	cat	NNS
kim	148	38	by	by	IN
kim	148	39	itself	-PRON-	PRP
kim	148	40	.	.	.
kim	149	1	See	see	VB
kim	149	2	Markoff	Markoff	NNP
kim	149	3	2012	2012	CD
kim	149	4	and	and	CC
kim	149	5	Liat	Liat	NNP
kim	149	6	2012	2012	CD
kim	149	7	.	.	.
kim	149	8	]	]	-RRB-
kim	150	1	[	[	-LRB-
kim	150	2	14	14	CD
kim	150	3	:	:	:
kim	150	4	This	this	DT
kim	150	5	black	black	JJ
kim	150	6	-	-	HYPH
kim	150	7	box	box	NN
kim	150	8	nature	nature	NN
kim	150	9	of	of	IN
kim	150	10	AI	AI	NNP
kim	150	11	systems	system	NNS
kim	150	12	powered	power	VBN
kim	150	13	by	by	IN
kim	150	14	machine	machine	NN
kim	150	15	learning	learning	NN
kim	150	16	has	have	VBZ
kim	150	17	raised	raise	VBN
kim	150	18	great	great	JJ
kim	150	19	concern	concern	NN
kim	150	20	among	among	IN
kim	150	21	many	many	JJ
kim	150	22	AI	AI	NNP
kim	150	23	researchers	researcher	NNS
kim	150	24	in	in	IN
kim	150	25	recent	recent	JJ
kim	150	26	years	year	NNS
kim	150	27	.	.	.
kim	151	1	This	this	DT
kim	151	2	is	be	VBZ
kim	151	3	problematic	problematic	JJ
kim	151	4	in	in	IN
kim	151	5	all	all	DT
kim	151	6	areas	area	NNS
kim	151	7	where	where	WRB
kim	151	8	these	these	DT
kim	151	9	AI	AI	NNP
kim	151	10	systems	system	NNS
kim	151	11	are	be	VBP
kim	151	12	used	use	VBN
kim	151	13	for	for	IN
kim	151	14	decision	decision	NN
kim	151	15	-	-	HYPH
kim	151	16	making	making	NN
kim	151	17	,	,	,
kim	151	18	not	not	RB
kim	151	19	just	just	RB
kim	151	20	in	in	IN
kim	151	21	military	military	JJ
kim	151	22	operations	operation	NNS
kim	151	23	.	.	.
kim	152	1	The	the	DT
kim	152	2	gravity	gravity	NN
kim	152	3	of	of	IN
kim	152	4	decisions	decision	NNS
kim	152	5	made	make	VBN
kim	152	6	in	in	IN
kim	152	7	a	a	DT
kim	152	8	military	military	JJ
kim	152	9	operation	operation	NN
kim	152	10	makes	make	VBZ
kim	152	11	this	this	DT
kim	152	12	problem	problem	NN
kim	152	13	even	even	RB
kim	152	14	more	more	RBR
kim	152	15	troublesome	troublesome	JJ
kim	152	16	.	.	.
kim	153	1	Fortunately	fortunately	RB
kim	153	2	,	,	,
kim	153	3	some	some	DT
kim	153	4	AI	AI	NNP
kim	153	5	researchers	researcher	NNS
kim	153	6	including	include	VBG
kim	153	7	those	those	DT
kim	153	8	in	in	IN
kim	153	9	the	the	DT
kim	153	10	US	US	NNP
kim	153	11	Department	Department	NNP
kim	153	12	of	of	IN
kim	153	13	Defense	Defense	NNP
kim	153	14	are	be	VBP
kim	153	15	actively	actively	RB
kim	153	16	working	work	VBG
kim	153	17	to	to	TO
kim	153	18	make	make	VB
kim	153	19	AI	AI	NNP
kim	153	20	systems	system	NNS
kim	153	21	explainable	explainable	JJ
kim	153	22	.	.	.
kim	154	1	But	but	CC
kim	154	2	until	until	IN
kim	154	3	such	such	JJ
kim	154	4	research	research	NN
kim	154	5	bears	bear	VBZ
kim	154	6	fruit	fruit	NN
kim	154	7	and	and	CC
kim	154	8	AI	AI	NNP
kim	154	9	systems	system	NNS
kim	154	10	become	become	VBP
kim	154	11	fully	fully	RB
kim	154	12	explainable	explainable	JJ
kim	154	13	,	,	,
kim	154	14	their	-PRON-	PRP$
kim	154	15	military	military	JJ
kim	154	16	use	use	NN
kim	154	17	means	mean	VBZ
kim	154	18	accepting	accept	VBG
kim	154	19	many	many	JJ
kim	154	20	unknown	unknown	JJ
kim	154	21	variables	variable	NNS
kim	154	22	and	and	CC
kim	154	23	unforeseeable	unforeseeable	JJ
kim	154	24	consequences	consequence	NNS
kim	154	25	.	.	.
kim	155	1	See	see	VB
kim	155	2	Turek	Turek	NNP
kim	155	3	n.d	n.d	NNP
kim	155	4	..	..	NFP
kim	155	5	]	]	-RRB-
kim	155	6	5	5	CD
kim	155	7	.	.	.
kim	156	1	AI	AI	NNP
kim	156	2	Applications	Applications	NNPS
kim	156	3	for	for	IN
kim	156	4	Libraries	library	NNS
kim	156	5	Do	do	VBP
kim	156	6	these	these	DT
kim	156	7	ethical	ethical	JJ
kim	156	8	concerns	concern	NNS
kim	156	9	outlined	outline	VBN
kim	156	10	above	above	IN
kim	156	11	apply	apply	VBP
kim	156	12	to	to	IN
kim	156	13	libraries	library	NNS
kim	156	14	?	?	.
kim	157	1	To	to	TO
kim	157	2	answer	answer	VB
kim	157	3	that	that	DT
kim	157	4	,	,	,
kim	157	5	let	let	VB
kim	157	6	us	-PRON-	PRP
kim	157	7	first	first	RB
kim	157	8	take	take	VB
kim	157	9	a	a	DT
kim	157	10	look	look	NN
kim	157	11	at	at	IN
kim	157	12	how	how	WRB
kim	157	13	AI	AI	NNP
kim	157	14	,	,	,
kim	157	15	particularly	particularly	RB
kim	157	16	machine	machine	NN
kim	157	17	learning	learning	NN
kim	157	18	,	,	,
kim	157	19	may	may	MD
kim	157	20	apply	apply	VB
kim	157	21	to	to	IN
kim	157	22	library	library	NN
kim	157	23	services	service	NNS
kim	157	24	and	and	CC
kim	157	25	operations	operation	NNS
kim	157	26	.	.	.
kim	158	1	AI	AI	NNP
kim	158	2	-	-	HYPH
kim	158	3	powered	power	VBN
kim	158	4	digital	digital	JJ
kim	158	5	assistants	assistant	NNS
kim	158	6	are	be	VBP
kim	158	7	likely	likely	JJ
kim	158	8	to	to	TO
kim	158	9	mediate	mediate	VB
kim	158	10	a	a	DT
kim	158	11	library	library	NN
kim	158	12	user	user	NN
kim	158	13	’s	’s	POS
kim	158	14	information	information	NN
kim	158	15	search	search	NN
kim	158	16	,	,	,
kim	158	17	discovery	discovery	NN
kim	158	18	,	,	,
kim	158	19	and	and	CC
kim	158	20	retrieval	retrieval	NN
kim	158	21	activities	activity	NNS
kim	158	22	in	in	IN
kim	158	23	the	the	DT
kim	158	24	near	near	JJ
kim	158	25	future	future	NN
kim	158	26	.	.	.
kim	159	1	In	in	IN
kim	159	2	recent	recent	JJ
kim	159	3	years	year	NNS
kim	159	4	,	,	,
kim	159	5	machine	machine	NN
kim	159	6	learning	learning	NN
kim	159	7	and	and	CC
kim	159	8	deep	deep	JJ
kim	159	9	learning	learning	NN
kim	159	10	have	have	VBP
kim	159	11	brought	bring	VBN
kim	159	12	significant	significant	JJ
kim	159	13	improvement	improvement	NN
kim	159	14	to	to	IN
kim	159	15	natural	natural	JJ
kim	159	16	language	language	NN
kim	159	17	processing	processing	NN
kim	159	18	(	(	-LRB-
kim	159	19	NLP	NLP	NNP
kim	159	20	)	)	-RRB-
kim	159	21	,	,	,
kim	159	22	which	which	WDT
kim	159	23	deals	deal	VBZ
kim	159	24	with	with	IN
kim	159	25	analyzing	analyze	VBG
kim	159	26	large	large	JJ
kim	159	27	amounts	amount	NNS
kim	159	28	of	of	IN
kim	159	29	natural	natural	JJ
kim	159	30	language	language	NN
kim	159	31	data	datum	NNS
kim	159	32	to	to	TO
kim	159	33	make	make	VB
kim	159	34	the	the	DT
kim	159	35	interaction	interaction	NN
kim	159	36	between	between	IN
kim	159	37	people	people	NNS
kim	159	38	and	and	CC
kim	159	39	machines	machine	NNS
kim	159	40	in	in	IN
kim	159	41	natural	natural	JJ
kim	159	42	languages	language	NNS
kim	159	43	possible	possible	JJ
kim	159	44	.	.	.
kim	160	1	For	for	IN
kim	160	2	instance	instance	NN
kim	160	3	,	,	,
kim	160	4	Google	Google	NNP
kim	160	5	Assistant	Assistant	NNP
kim	160	6	’s	’s	POS
kim	160	7	new	new	JJ
kim	160	8	feature	feature	NN
kim	160	9	‘	'	``
kim	160	10	duplex	duplex	NN
kim	160	11	’	'	''
kim	160	12	was	be	VBD
kim	160	13	shown	show	VBN
kim	160	14	to	to	TO
kim	160	15	successfully	successfully	RB
kim	160	16	make	make	VB
kim	160	17	a	a	DT
kim	160	18	phone	phone	NN
kim	160	19	reservation	reservation	NN
kim	160	20	with	with	IN
kim	160	21	restaurant	restaurant	NN
kim	160	22	staff	staff	NN
kim	160	23	in	in	IN
kim	160	24	2018	2018	CD
kim	160	25	(	(	-LRB-
kim	160	26	Welch	Welch	NNP
kim	160	27	2018	2018	CD
kim	160	28	)	)	-RRB-
kim	160	29	.	.	.
kim	161	1	Google	Google	NNP
kim	161	2	’s	’s	POS
kim	161	3	real	real	JJ
kim	161	4	-	-	HYPH
kim	161	5	time	time	NN
kim	161	6	translation	translation	NN
kim	161	7	capability	capability	NN
kim	161	8	for	for	IN
kim	161	9	40	40	CD
kim	161	10	different	different	JJ
kim	161	11	languages	language	NNS
kim	161	12	is	be	VBZ
kim	161	13	planned	plan	VBN
kim	161	14	to	to	TO
kim	161	15	become	become	VB
kim	161	16	part	part	NN
kim	161	17	of	of	IN
kim	161	18	the	the	DT
kim	161	19	features	feature	NNS
kim	161	20	for	for	IN
kim	161	21	all	all	DT
kim	161	22	headphones	headphone	NNS
kim	161	23	equipped	equip	VBN
kim	161	24	with	with	IN
kim	161	25	Google	Google	NNP
kim	161	26	Assistant	Assistant	NNP
kim	161	27	(	(	-LRB-
kim	161	28	Le	Le	NNP
kim	161	29	and	and	CC
kim	161	30	Schuster	Schuster	NNP
kim	161	31	2016	2016	CD
kim	161	32	,	,	,
kim	161	33	Porter	Porter	NNP
kim	161	34	2018	2018	CD
kim	161	35	)	)	-RRB-
kim	161	36	.	.	.
kim	162	1	As	as	IN
kim	162	2	digital	digital	JJ
kim	162	3	assistants	assistant	NNS
kim	162	4	become	become	VBP
kim	162	5	capable	capable	JJ
kim	162	6	of	of	IN
kim	162	7	handling	handle	VBG
kim	162	8	more	more	RBR
kim	162	9	sophisticated	sophisticated	JJ
kim	162	10	language	language	NN
kim	162	11	tasks	task	NNS
kim	162	12	,	,	,
kim	162	13	their	-PRON-	PRP$
kim	162	14	use	use	NN
kim	162	15	as	as	IN
kim	162	16	a	a	DT
kim	162	17	flexible	flexible	JJ
kim	162	18	voice	voice	NN
kim	162	19	user	user	NN
kim	162	20	interface	interface	NN
kim	162	21	will	will	MD
kim	162	22	only	only	RB
kim	162	23	increase	increase	VB
kim	162	24	.	.	.
kim	163	1	Such	such	JJ
kim	163	2	digital	digital	JJ
kim	163	3	assistants	assistant	NNS
kim	163	4	will	will	MD
kim	163	5	directly	directly	RB
kim	163	6	interact	interact	VB
kim	163	7	with	with	IN
kim	163	8	library	library	JJ
kim	163	9	systems	system	NNS
kim	163	10	and	and	CC
kim	163	11	applications	application	NNS
kim	163	12	,	,	,
kim	163	13	automatically	automatically	RB
kim	163	14	interpret	interpret	VB
kim	163	15	a	a	DT
kim	163	16	query	query	NN
kim	163	17	,	,	,
kim	163	18	and	and	CC
kim	163	19	return	return	VB
kim	163	20	results	result	NNS
kim	163	21	that	that	WDT
kim	163	22	they	-PRON-	PRP
kim	163	23	deem	deem	VBP
kim	163	24	to	to	TO
kim	163	25	be	be	VB
kim	163	26	most	most	RBS
kim	163	27	relevant	relevant	JJ
kim	163	28	.	.	.
kim	164	1	Those	those	DT
kim	164	2	digital	digital	JJ
kim	164	3	assistants	assistant	NNS
kim	164	4	can	can	MD
kim	164	5	also	also	RB
kim	164	6	be	be	VB
kim	164	7	equipped	equip	VBN
kim	164	8	to	to	TO
kim	164	9	handle	handle	VB
kim	164	10	the	the	DT
kim	164	11	library	library	NN
kim	164	12	’s	’s	,
kim	164	13	traditional	traditional	JJ
kim	164	14	reference	reference	NN
kim	164	15	or	or	CC
kim	164	16	readers	reader	NNS
kim	164	17	’	’	POS
kim	164	18	advisory	advisory	JJ
kim	164	19	service	service	NN
kim	164	20	.	.	.
kim	165	1	Integrated	integrate	VBN
kim	165	2	into	into	IN
kim	165	3	a	a	DT
kim	165	4	humanoid	humanoid	JJ
kim	165	5	robot	robot	NN
kim	165	6	body	body	NN
kim	165	7	,	,	,
kim	165	8	they	-PRON-	PRP
kim	165	9	may	may	MD
kim	165	10	even	even	RB
kim	165	11	greet	greet	VB
kim	165	12	library	library	JJ
kim	165	13	patrons	patron	NNS
kim	165	14	at	at	IN
kim	165	15	the	the	DT
kim	165	16	entrance	entrance	NN
kim	165	17	and	and	CC
kim	165	18	answer	answer	VB
kim	165	19	directional	directional	JJ
kim	165	20	questions	question	NNS
kim	165	21	about	about	IN
kim	165	22	the	the	DT
kim	165	23	library	library	NN
kim	165	24	building	building	NN
kim	165	25	.	.	.
kim	166	1	Cataloging	catalog	VBG
kim	166	2	,	,	,
kim	166	3	abstracting	abstracting	NN
kim	166	4	,	,	,
kim	166	5	and	and	CC
kim	166	6	indexing	indexing	NN
kim	166	7	are	be	VBP
kim	166	8	other	other	JJ
kim	166	9	areas	area	NNS
kim	166	10	where	where	WRB
kim	166	11	AI	AI	NNP
kim	166	12	will	will	MD
kim	166	13	be	be	VB
kim	166	14	actively	actively	RB
kim	166	15	utilized	utilize	VBN
kim	166	16	.	.	.
kim	167	1	Currently	currently	RB
kim	167	2	,	,	,
kim	167	3	those	those	DT
kim	167	4	tasks	task	NNS
kim	167	5	are	be	VBP
kim	167	6	performed	perform	VBN
kim	167	7	by	by	IN
kim	167	8	skilled	skilled	JJ
kim	167	9	professionals	professional	NNS
kim	167	10	.	.	.
kim	168	1	But	but	CC
kim	168	2	as	as	IN
kim	168	3	AI	AI	NNP
kim	168	4	applications	application	NNS
kim	168	5	become	become	VBP
kim	168	6	more	more	RBR
kim	168	7	sophisticated	sophisticated	JJ
kim	168	8	,	,	,
kim	168	9	we	-PRON-	PRP
kim	168	10	may	may	MD
kim	168	11	see	see	VB
kim	168	12	many	many	JJ
kim	168	13	of	of	IN
kim	168	14	those	those	DT
kim	168	15	tasks	task	NNS
kim	168	16	partially	partially	RB
kim	168	17	or	or	CC
kim	168	18	fully	fully	RB
kim	168	19	automated	automate	VBN
kim	168	20	and	and	CC
kim	168	21	handed	hand	VBD
kim	168	22	over	over	RP
kim	168	23	to	to	IN
kim	168	24	AI	AI	NNP
kim	168	25	systems	system	NNS
kim	168	26	.	.	.
kim	169	1	Machine	machine	NN
kim	169	2	learning	learning	NN
kim	169	3	and	and	CC
kim	169	4	deep	deep	JJ
kim	169	5	learning	learning	NN
kim	169	6	can	can	MD
kim	169	7	be	be	VB
kim	169	8	used	use	VBN
kim	169	9	to	to	TO
kim	169	10	extract	extract	VB
kim	169	11	key	key	JJ
kim	169	12	information	information	NN
kim	169	13	from	from	IN
kim	169	14	a	a	DT
kim	169	15	large	large	JJ
kim	169	16	number	number	NN
kim	169	17	of	of	IN
kim	169	18	documents	document	NNS
kim	169	19	or	or	CC
kim	169	20	from	from	IN
kim	169	21	information	information	NN
kim	169	22	-	-	HYPH
kim	169	23	rich	rich	JJ
kim	169	24	visual	visual	JJ
kim	169	25	materials	material	NNS
kim	169	26	such	such	JJ
kim	169	27	as	as	IN
kim	169	28	maps	map	NNS
kim	169	29	and	and	CC
kim	169	30	video	video	NN
kim	169	31	recordings	recording	NNS
kim	169	32	,	,	,
kim	169	33	and	and	CC
kim	169	34	generate	generate	VB
kim	169	35	metadata	metadata	NN
kim	169	36	or	or	CC
kim	169	37	a	a	DT
kim	169	38	summary	summary	NN
kim	169	39	.	.	.
kim	170	1	Since	since	IN
kim	170	2	machine	machine	NN
kim	170	3	learning	learning	NN
kim	170	4	is	be	VBZ
kim	170	5	new	new	JJ
kim	170	6	to	to	IN
kim	170	7	libraries	library	NNS
kim	170	8	,	,	,
kim	170	9	there	there	EX
kim	170	10	are	be	VBP
kim	170	11	a	a	DT
kim	170	12	relatively	relatively	RB
kim	170	13	small	small	JJ
kim	170	14	number	number	NN
kim	170	15	of	of	IN
kim	170	16	machine	machine	NN
kim	170	17	learning	learning	NN
kim	170	18	applications	application	NNS
kim	170	19	developed	develop	VBN
kim	170	20	for	for	IN
kim	170	21	libraries	library	NNS
kim	170	22	’	’	POS
kim	170	23	use	use	NN
kim	170	24	.	.	.
kim	171	1	They	-PRON-	PRP
kim	171	2	are	be	VBP
kim	171	3	likely	likely	JJ
kim	171	4	to	to	TO
kim	171	5	grow	grow	VB
kim	171	6	in	in	IN
kim	171	7	number	number	NN
kim	171	8	.	.	.
kim	172	1	Yewno	Yewno	NNP
kim	172	2	,	,	,
kim	172	3	Quartolio	Quartolio	NNP
kim	172	4	,	,	,
kim	172	5	and	and	CC
kim	172	6	Iris.ai	iris.ai	NN
kim	172	7	are	be	VBP
kim	172	8	examples	example	NNS
kim	172	9	of	of	IN
kim	172	10	the	the	DT
kim	172	11	vendor	vendor	NN
kim	172	12	products	product	NNS
kim	172	13	developed	develop	VBN
kim	172	14	with	with	IN
kim	172	15	machine	machine	NN
kim	172	16	learning	learning	NN
kim	172	17	and	and	CC
kim	172	18	deep	deep	JJ
kim	172	19	learning	learning	NN
kim	172	20	techniques	technique	NNS
kim	172	21	.	.	.
kim	173	1	Yewno	Yewno	NNP
kim	173	2	Discover	Discover	NNP
kim	173	3	displays	display	VBZ
kim	173	4	the	the	DT
kim	173	5	connections	connection	NNS
kim	173	6	between	between	IN
kim	173	7	different	different	JJ
kim	173	8	concepts	concept	NNS
kim	173	9	or	or	CC
kim	173	10	works	work	NNS
kim	173	11	in	in	IN
kim	173	12	library	library	NN
kim	173	13	materials	material	NNS
kim	173	14	.	.	.
kim	174	1	Quartolio	quartolio	JJ
kim	174	2	targets	target	NNS
kim	174	3	researchers	researcher	NNS
kim	174	4	looking	look	VBG
kim	174	5	to	to	TO
kim	174	6	discover	discover	VB
kim	174	7	untapped	untapped	JJ
kim	174	8	research	research	NN
kim	174	9	opportunities	opportunity	NNS
kim	174	10	based	base	VBN
kim	174	11	upon	upon	IN
kim	174	12	a	a	DT
kim	174	13	large	large	JJ
kim	174	14	amount	amount	NN
kim	174	15	of	of	IN
kim	174	16	data	datum	NNS
kim	174	17	that	that	WDT
kim	174	18	includes	include	VBZ
kim	174	19	articles	article	NNS
kim	174	20	,	,	,
kim	174	21	clinical	clinical	JJ
kim	174	22	trials	trial	NNS
kim	174	23	,	,	,
kim	174	24	patents	patent	NNS
kim	174	25	,	,	,
kim	174	26	and	and	CC
kim	174	27	notes	note	NNS
kim	174	28	.	.	.
kim	175	1	Similarly	similarly	RB
kim	175	2	,	,	,
kim	175	3	Iris.ai	Iris.ai	NNP
kim	175	4	helps	help	VBZ
kim	175	5	researchers	researcher	NNS
kim	175	6	identify	identify	VB
kim	175	7	and	and	CC
kim	175	8	review	review	VB
kim	175	9	a	a	DT
kim	175	10	large	large	JJ
kim	175	11	amount	amount	NN
kim	175	12	of	of	IN
kim	175	13	research	research	NN
kim	175	14	papers	paper	NNS
kim	175	15	and	and	CC
kim	175	16	patents	patent	NNS
kim	175	17	and	and	CC
kim	175	18	extracts	extract	VBZ
kim	175	19	key	key	JJ
kim	175	20	information	information	NN
kim	175	21	from	from	IN
kim	175	22	them	-PRON-	PRP
kim	175	23	.	.	.
kim	176	1	[	[	-LRB-
kim	176	2	footnoteRef:15	footnoteref:15	CD
kim	176	3	]	]	-RRB-
kim	176	4	Kira	Kira	NNP
kim	176	5	identifies	identifie	NNS
kim	176	6	,	,	,
kim	176	7	extracts	extract	VBZ
kim	176	8	,	,	,
kim	176	9	and	and	CC
kim	176	10	analyzes	analyze	VBZ
kim	176	11	text	text	NN
kim	176	12	in	in	IN
kim	176	13	contracts	contract	NNS
kim	176	14	and	and	CC
kim	176	15	other	other	JJ
kim	176	16	legal	legal	JJ
kim	176	17	documents	document	NNS
kim	176	18	.	.	.
kim	177	1	[	[	-LRB-
kim	177	2	footnoteRef:16	footnoteRef:16	NNP
kim	177	3	]	]	-RRB-
kim	177	4	None	none	NN
kim	177	5	of	of	IN
kim	177	6	these	these	DT
kim	177	7	applications	application	NNS
kim	177	8	performs	perform	VBZ
kim	177	9	fully	fully	RB
kim	177	10	automated	automate	VBN
kim	177	11	decision	decision	NN
kim	177	12	-	-	HYPH
kim	177	13	making	making	NN
kim	177	14	nor	nor	CC
kim	177	15	incorporates	incorporate	VBZ
kim	177	16	the	the	DT
kim	177	17	digital	digital	JJ
kim	177	18	assistant	assistant	NN
kim	177	19	feature	feature	NN
kim	177	20	.	.	.
kim	178	1	But	but	CC
kim	178	2	this	this	DT
kim	178	3	is	be	VBZ
kim	178	4	an	an	DT
kim	178	5	area	area	NN
kim	178	6	where	where	WRB
kim	178	7	information	information	NN
kim	178	8	systems	system	NNS
kim	178	9	vendors	vendor	NNS
kim	178	10	are	be	VBP
kim	178	11	increasingly	increasingly	RB
kim	178	12	focusing	focus	VBG
kim	178	13	.	.	.
kim	179	1	[	[	-LRB-
kim	179	2	15	15	CD
kim	179	3	:	:	:
kim	179	4	See	see	VB
kim	179	5	https://www.yewno.com/education	https://www.yewno.com/education	CD
kim	179	6	,	,	,
kim	179	7	https://quartolio.com/	https://quartolio.com/	CD
kim	179	8	,	,	,
kim	179	9	and	and	CC
kim	179	10	https://iris.ai/.	https://iris.ai/.	VBN
kim	179	11	]	]	-RRB-
kim	180	1	[	[	-LRB-
kim	180	2	16	16	CD
kim	180	3	:	:	:
kim	180	4	See	see	VB
kim	180	5	https://kirasystems.com/.	https://kirasystems.com/.	NNP
kim	181	1	Law	law	NN
kim	181	2	firms	firm	NNS
kim	181	3	are	be	VBP
kim	181	4	adopting	adopt	VBG
kim	181	5	similar	similar	JJ
kim	181	6	products	product	NNS
kim	181	7	to	to	TO
kim	181	8	automate	automate	VB
kim	181	9	and	and	CC
kim	181	10	expedite	expedite	VB
kim	181	11	their	-PRON-	PRP$
kim	181	12	legal	legal	JJ
kim	181	13	work	work	NN
kim	181	14	,	,	,
kim	181	15	and	and	CC
kim	181	16	law	law	NN
kim	181	17	librarians	librarian	NNS
kim	181	18	are	be	VBP
kim	181	19	discussing	discuss	VBG
kim	181	20	how	how	WRB
kim	181	21	the	the	DT
kim	181	22	use	use	NN
kim	181	23	of	of	IN
kim	181	24	AI	AI	NNP
kim	181	25	may	may	MD
kim	181	26	change	change	VB
kim	181	27	their	-PRON-	PRP$
kim	181	28	work	work	NN
kim	181	29	.	.	.
kim	182	1	See	see	VB
kim	182	2	Marr	Marr	NNP
kim	182	3	2018	2018	CD
kim	182	4	and	and	CC
kim	182	5	Talley	Talley	NNP
kim	182	6	2016	2016	CD
kim	182	7	.	.	.
kim	182	8	]	]	-RRB-
kim	183	1	Libraries	library	NNS
kim	183	2	themselves	-PRON-	PRP
kim	183	3	are	be	VBP
kim	183	4	also	also	RB
kim	183	5	experimenting	experiment	VBG
kim	183	6	with	with	IN
kim	183	7	AI	AI	NNP
kim	183	8	to	to	TO
kim	183	9	test	test	VB
kim	183	10	its	-PRON-	PRP$
kim	183	11	potential	potential	NN
kim	183	12	for	for	IN
kim	183	13	library	library	NN
kim	183	14	services	service	NNS
kim	183	15	and	and	CC
kim	183	16	operations	operation	NNS
kim	183	17	.	.	.
kim	184	1	Some	some	DT
kim	184	2	are	be	VBP
kim	184	3	focusing	focus	VBG
kim	184	4	on	on	IN
kim	184	5	using	use	VBG
kim	184	6	AI	AI	NNP
kim	184	7	,	,	,
kim	184	8	particularly	particularly	RB
kim	184	9	the	the	DT
kim	184	10	voice	voice	NN
kim	184	11	user	user	NN
kim	184	12	interface	interface	NN
kim	184	13	aspect	aspect	NN
kim	184	14	of	of	IN
kim	184	15	the	the	DT
kim	184	16	digital	digital	JJ
kim	184	17	assistant	assistant	NN
kim	184	18	,	,	,
kim	184	19	in	in	IN
kim	184	20	order	order	NN
kim	184	21	to	to	TO
kim	184	22	improve	improve	VB
kim	184	23	existing	exist	VBG
kim	184	24	services	service	NNS
kim	184	25	.	.	.
kim	185	1	The	the	DT
kim	185	2	University	University	NNP
kim	185	3	of	of	IN
kim	185	4	Oklahoma	Oklahoma	NNP
kim	185	5	Libraries	library	NNS
kim	185	6	have	have	VBP
kim	185	7	been	be	VBN
kim	185	8	building	build	VBG
kim	185	9	an	an	DT
kim	185	10	Alexa	alexa	JJ
kim	185	11	application	application	NN
kim	185	12	to	to	TO
kim	185	13	provide	provide	VB
kim	185	14	basic	basic	JJ
kim	185	15	reference	reference	NN
kim	185	16	service	service	NN
kim	185	17	to	to	IN
kim	185	18	their	-PRON-	PRP$
kim	185	19	students	student	NNS
kim	185	20	.	.	.
kim	186	1	[	[	-LRB-
kim	186	2	footnoteRef:17	footnoteRef:17	NNP
kim	186	3	]	]	-RRB-
kim	186	4	At	at	IN
kim	186	5	the	the	DT
kim	186	6	University	University	NNP
kim	186	7	of	of	IN
kim	186	8	Pretoria	Pretoria	NNP
kim	186	9	Library	Library	NNP
kim	186	10	in	in	IN
kim	186	11	South	South	NNP
kim	186	12	Africa	Africa	NNP
kim	186	13	,	,	,
kim	186	14	a	a	DT
kim	186	15	robot	robot	NN
kim	186	16	named	name	VBN
kim	186	17	‘	'	``
kim	186	18	Libby	Libby	NNP
kim	186	19	’	'	''
kim	186	20	already	already	RB
kim	186	21	interacts	interact	VBZ
kim	186	22	with	with	IN
kim	186	23	patrons	patron	NNS
kim	186	24	by	by	IN
kim	186	25	providing	provide	VBG
kim	186	26	guidance	guidance	NN
kim	186	27	,	,	,
kim	186	28	answering	answer	VBG
kim	186	29	questions	question	NNS
kim	186	30	,	,	,
kim	186	31	conducting	conduct	VBG
kim	186	32	surveys	survey	NNS
kim	186	33	,	,	,
kim	186	34	and	and	CC
kim	186	35	displaying	display	VBG
kim	186	36	marketing	marketing	NN
kim	186	37	videos	video	NNS
kim	186	38	(	(	-LRB-
kim	186	39	Mahlangu	Mahlangu	NNP
kim	186	40	2019	2019	CD
kim	186	41	)	)	-RRB-
kim	186	42	.	.	.
kim	187	1	[	[	-LRB-
kim	187	2	17	17	CD
kim	187	3	:	:	:
kim	187	4	University	University	NNP
kim	187	5	of	of	IN
kim	187	6	Oklahoma	Oklahoma	NNP
kim	187	7	Libraries	Libraries	NNPS
kim	187	8	are	be	VBP
kim	187	9	building	build	VBG
kim	187	10	an	an	DT
kim	187	11	Alexa	alexa	JJ
kim	187	12	application	application	NN
kim	187	13	that	that	WDT
kim	187	14	will	will	MD
kim	187	15	provide	provide	VB
kim	187	16	some	some	DT
kim	187	17	basic	basic	JJ
kim	187	18	reference	reference	NN
kim	187	19	service	service	NN
kim	187	20	to	to	IN
kim	187	21	their	-PRON-	PRP$
kim	187	22	students	student	NNS
kim	187	23	.	.	.
kim	188	1	Also	also	RB
kim	188	2	,	,	,
kim	188	3	their	-PRON-	PRP$
kim	188	4	PAIR	PAIR	NNP
kim	188	5	registry	registry	NN
kim	188	6	attempts	attempt	VBZ
kim	188	7	to	to	TO
kim	188	8	compile	compile	VB
kim	188	9	all	all	DT
kim	188	10	AI	ai	NN
kim	188	11	-	-	HYPH
kim	188	12	related	relate	VBN
kim	188	13	projects	project	NNS
kim	188	14	at	at	IN
kim	188	15	libraries	library	NNS
kim	188	16	.	.	.
kim	189	1	See	see	VB
kim	189	2	https://pair.libraries.ou.edu	https://pair.libraries.ou.edu	NN
kim	189	3	.	.	.
kim	189	4	]	]	-RRB-
kim	190	1	Other	other	JJ
kim	190	2	libraries	library	NNS
kim	190	3	are	be	VBP
kim	190	4	applying	apply	VBG
kim	190	5	AI	AI	NNP
kim	190	6	to	to	IN
kim	190	7	extracting	extract	VBG
kim	190	8	information	information	NN
kim	190	9	and	and	CC
kim	190	10	automating	automate	VBG
kim	190	11	metadata	metadata	NN
kim	190	12	generation	generation	NN
kim	190	13	for	for	IN
kim	190	14	digital	digital	JJ
kim	190	15	materials	material	NNS
kim	190	16	to	to	TO
kim	190	17	enhance	enhance	VB
kim	190	18	their	-PRON-	PRP$
kim	190	19	discovery	discovery	NN
kim	190	20	and	and	CC
kim	190	21	use	use	NN
kim	190	22	.	.	.
kim	191	1	The	the	DT
kim	191	2	Library	Library	NNP
kim	191	3	of	of	IN
kim	191	4	Congress	Congress	NNP
kim	191	5	has	have	VBZ
kim	191	6	worked	work	VBN
kim	191	7	on	on	IN
kim	191	8	detecting	detect	VBG
kim	191	9	features	feature	NNS
kim	191	10	,	,	,
kim	191	11	such	such	JJ
kim	191	12	as	as	IN
kim	191	13	railroads	railroad	NNS
kim	191	14	in	in	IN
kim	191	15	maps	map	NNS
kim	191	16	,	,	,
kim	191	17	using	use	VBG
kim	191	18	the	the	DT
kim	191	19	convolutional	convolutional	JJ
kim	191	20	neural	neural	JJ
kim	191	21	network	network	NN
kim	191	22	model	model	NN
kim	191	23	,	,	,
kim	191	24	and	and	CC
kim	191	25	issued	issue	VBD
kim	191	26	a	a	DT
kim	191	27	solicitation	solicitation	NN
kim	191	28	for	for	IN
kim	191	29	a	a	DT
kim	191	30	machine	machine	NN
kim	191	31	learning	learning	NN
kim	191	32	and	and	CC
kim	191	33	deep	deep	JJ
kim	191	34	learning	learning	NN
kim	191	35	pilot	pilot	NN
kim	191	36	program	program	NN
kim	191	37	that	that	WDT
kim	191	38	will	will	MD
kim	191	39	maximize	maximize	VB
kim	191	40	the	the	DT
kim	191	41	use	use	NN
kim	191	42	of	of	IN
kim	191	43	its	-PRON-	PRP$
kim	191	44	digital	digital	JJ
kim	191	45	collections	collection	NNS
kim	191	46	in	in	IN
kim	191	47	2019	2019	CD
kim	191	48	.	.	.
kim	192	1	[	[	-LRB-
kim	192	2	footnoteRef:18	footnoteRef:18	NNP
kim	192	3	]	]	-RRB-
kim	192	4	Indiana	Indiana	NNP
kim	192	5	University	University	NNP
kim	192	6	Libraries	Libraries	NNPS
kim	192	7	,	,	,
kim	192	8	AVP	AVP	NNP
kim	192	9	,	,	,
kim	192	10	University	University	NNP
kim	192	11	of	of	IN
kim	192	12	Texas	Texas	NNP
kim	192	13	Austin	Austin	NNP
kim	192	14	School	School	NNP
kim	192	15	of	of	IN
kim	192	16	Information	Information	NNP
kim	192	17	,	,	,
kim	192	18	and	and	CC
kim	192	19	the	the	DT
kim	192	20	New	New	NNP
kim	192	21	York	York	NNP
kim	192	22	Public	Public	NNP
kim	192	23	Library	Library	NNP
kim	192	24	are	be	VBP
kim	192	25	jointly	jointly	RB
kim	192	26	developing	develop	VBG
kim	192	27	the	the	DT
kim	192	28	Audiovisual	Audiovisual	NNP
kim	192	29	Metadata	Metadata	NNP
kim	192	30	Platform	Platform	NNP
kim	192	31	(	(	-LRB-
kim	192	32	AMP	AMP	NNP
kim	192	33	)	)	-RRB-
kim	192	34	,	,	,
kim	192	35	using	use	VBG
kim	192	36	many	many	JJ
kim	192	37	AI	ai	NN
kim	192	38	tools	tool	NNS
kim	192	39	in	in	IN
kim	192	40	order	order	NN
kim	192	41	to	to	TO
kim	192	42	automatically	automatically	RB
kim	192	43	generate	generate	VB
kim	192	44	metadata	metadata	NN
kim	192	45	for	for	IN
kim	192	46	audiovisual	audiovisual	JJ
kim	192	47	materials	material	NNS
kim	192	48	,	,	,
kim	192	49	which	which	WDT
kim	192	50	collection	collection	NN
kim	192	51	managers	manager	NNS
kim	192	52	can	can	MD
kim	192	53	use	use	VB
kim	192	54	to	to	TO
kim	192	55	supplement	supplement	VB
kim	192	56	their	-PRON-	PRP$
kim	192	57	archival	archival	NN
kim	192	58	description	description	NN
kim	192	59	and	and	CC
kim	192	60	processing	processing	NN
kim	192	61	workflows	workflow	NNS
kim	192	62	.	.	.
kim	193	1	[	[	-LRB-
kim	193	2	footnoteRef:19	footnoteref:19	NN
kim	193	3	]	]	-RRB-
kim	193	4	[	[	-LRB-
kim	193	5	18	18	CD
kim	193	6	:	:	:
kim	193	7	See	see	VB
kim	193	8	Blewer	Blewer	NNP
kim	193	9	,	,	,
kim	193	10	Kim	Kim	NNP
kim	193	11	,	,	,
kim	193	12	and	and	CC
kim	193	13	Phetteplace	Phetteplace	NNP
kim	193	14	2018	2018	CD
kim	193	15	and	and	CC
kim	193	16	Price	Price	NNP
kim	193	17	2019	2019	CD
kim	193	18	.	.	.
kim	193	19	]	]	-RRB-
kim	194	1	[	[	-LRB-
kim	194	2	19	19	CD
kim	194	3	:	:	:
kim	194	4	The	the	DT
kim	194	5	AMP	AMP	NNP
kim	194	6	wiki	wiki	NN
kim	194	7	is	be	VBZ
kim	194	8	https://wiki.dlib.indiana.edu/pages/viewpage.action?pageId=531699941	https://wiki.dlib.indiana.edu/pages/viewpage.action?pageid=531699941	NN
kim	194	9	.	.	.
kim	195	1	The	the	DT
kim	195	2	Audiovisual	audiovisual	JJ
kim	195	3	Metadata	metadata	NN
kim	195	4	Platform	platform	NN
kim	195	5	Pilot	pilot	NN
kim	195	6	Development	Development	NNP
kim	195	7	(	(	-LRB-
kim	195	8	AMPPD	AMPPD	NNP
kim	195	9	)	)	-RRB-
kim	195	10	project	project	NN
kim	195	11	was	be	VBD
kim	195	12	presented	present	VBN
kim	195	13	at	at	IN
kim	195	14	Code4Lib	Code4Lib	NNP
kim	195	15	2020	2020	CD
kim	195	16	(	(	-LRB-
kim	195	17	Averkamp	Averkamp	NNP
kim	195	18	and	and	CC
kim	195	19	Hardesty	Hardesty	NNP
kim	195	20	2020	2020	CD
kim	195	21	)	)	-RRB-
kim	195	22	.	.	.
kim	195	23	]	]	-RRB-
kim	196	1	Some	some	DT
kim	196	2	libraries	library	NNS
kim	196	3	are	be	VBP
kim	196	4	also	also	RB
kim	196	5	testing	test	VBG
kim	196	6	out	out	RP
kim	196	7	AI	AI	NNP
kim	196	8	as	as	IN
kim	196	9	a	a	DT
kim	196	10	tool	tool	NN
kim	196	11	for	for	IN
kim	196	12	evaluating	evaluate	VBG
kim	196	13	services	service	NNS
kim	196	14	and	and	CC
kim	196	15	operations	operation	NNS
kim	196	16	.	.	.
kim	197	1	The	the	DT
kim	197	2	University	University	NNP
kim	197	3	of	of	IN
kim	197	4	Rochester	Rochester	NNP
kim	197	5	Libraries	Libraries	NNP
kim	197	6	applied	apply	VBD
kim	197	7	deep	deep	JJ
kim	197	8	learning	learning	NN
kim	197	9	to	to	IN
kim	197	10	the	the	DT
kim	197	11	library	library	NN
kim	197	12	’s	’s	POS
kim	197	13	space	space	NN
kim	197	14	assessment	assessment	NN
kim	197	15	to	to	TO
kim	197	16	determine	determine	VB
kim	197	17	the	the	DT
kim	197	18	optimal	optimal	JJ
kim	197	19	staffing	staffing	NN
kim	197	20	level	level	NN
kim	197	21	and	and	CC
kim	197	22	building	building	NN
kim	197	23	hours	hour	NNS
kim	197	24	.	.	.
kim	198	1	The	the	DT
kim	198	2	University	University	NNP
kim	198	3	of	of	IN
kim	198	4	Illinois	Illinois	NNP
kim	198	5	Urbana	Urbana	NNP
kim	198	6	-	-	HYPH
kim	198	7	Champaign	Champaign	NNP
kim	198	8	Libraries	Libraries	NNPS
kim	198	9	used	use	VBD
kim	198	10	machine	machine	NN
kim	198	11	learning	learn	VBG
kim	198	12	to	to	TO
kim	198	13	conduct	conduct	VB
kim	198	14	sentiment	sentiment	NN
kim	198	15	analysis	analysis	NN
kim	198	16	on	on	IN
kim	198	17	their	-PRON-	PRP$
kim	198	18	reference	reference	NN
kim	198	19	chat	chat	NN
kim	198	20	log	log	NN
kim	198	21	(	(	-LRB-
kim	198	22	Blewer	Blewer	NNP
kim	198	23	,	,	,
kim	198	24	Kim	Kim	NNP
kim	198	25	,	,	,
kim	198	26	and	and	CC
kim	198	27	Phetteplace	Phetteplace	NNP
kim	198	28	2018	2018	CD
kim	198	29	)	)	-RRB-
kim	198	30	.	.	.
kim	199	1	6	6	CD
kim	199	2	.	.	.
kim	200	1	Ethical	ethical	JJ
kim	200	2	Challenges	Challenges	NNPS
kim	200	3	from	from	IN
kim	200	4	the	the	DT
kim	200	5	Personalized	Personalized	NNP
kim	200	6	and	and	CC
kim	200	7	Automated	Automated	NNP
kim	200	8	Information	Information	NNP
kim	200	9	Environment	Environment	NNP
kim	200	10	Do	do	VBP
kim	200	11	these	these	DT
kim	200	12	current	current	JJ
kim	200	13	and	and	CC
kim	200	14	future	future	JJ
kim	200	15	AI	AI	NNP
kim	200	16	applications	application	NNS
kim	200	17	for	for	IN
kim	200	18	libraries	library	NNS
kim	200	19	pose	pose	VBP
kim	200	20	ethical	ethical	JJ
kim	200	21	challenges	challenge	NNS
kim	200	22	similar	similar	JJ
kim	200	23	to	to	IN
kim	200	24	those	those	DT
kim	200	25	that	that	WDT
kim	200	26	we	-PRON-	PRP
kim	200	27	discussed	discuss	VBD
kim	200	28	earlier	early	RBR
kim	200	29	?	?	.
kim	201	1	Since	since	IN
kim	201	2	information	information	NN
kim	201	3	query	query	NN
kim	201	4	,	,	,
kim	201	5	discovery	discovery	NN
kim	201	6	,	,	,
kim	201	7	and	and	CC
kim	201	8	retrieval	retrieval	NN
kim	201	9	rarely	rarely	RB
kim	201	10	involve	involve	VBP
kim	201	11	life	life	NN
kim	201	12	-	-	HYPH
kim	201	13	or	or	CC
kim	201	14	-	-	HYPH
kim	201	15	death	death	NN
kim	201	16	situations	situation	NNS
kim	201	17	,	,	,
kim	201	18	stakes	stake	NNS
kim	201	19	seem	seem	VBP
kim	201	20	to	to	TO
kim	201	21	be	be	VB
kim	201	22	certainly	certainly	RB
kim	201	23	lower	low	JJR
kim	201	24	.	.	.
kim	202	1	But	but	CC
kim	202	2	an	an	DT
kim	202	3	AI	ai	NN
kim	202	4	-	-	HYPH
kim	202	5	driven	drive	VBN
kim	202	6	automated	automated	JJ
kim	202	7	information	information	NN
kim	202	8	environment	environment	NN
kim	202	9	does	do	VBZ
kim	202	10	raise	raise	VB
kim	202	11	its	-PRON-	PRP$
kim	202	12	own	own	JJ
kim	202	13	distinct	distinct	JJ
kim	202	14	ethical	ethical	JJ
kim	202	15	challenges	challenge	NNS
kim	202	16	.	.	.
kim	203	1	(	(	-LRB-
kim	203	2	i	i	NN
kim	203	3	)	)	-RRB-
kim	203	4	Intellectual	intellectual	JJ
kim	203	5	isolation	isolation	NN
kim	203	6	and	and	CC
kim	203	7	bigotry	bigotry	NN
kim	203	8	hampering	hamper	VBG
kim	203	9	civic	civic	JJ
kim	203	10	discourse	discourse	NN
kim	203	11	Many	many	JJ
kim	203	12	AI	AI	NNP
kim	203	13	applications	application	NNS
kim	203	14	that	that	WDT
kim	203	15	assist	assist	VBP
kim	203	16	with	with	IN
kim	203	17	information	information	NN
kim	203	18	seeking	seek	VBG
kim	203	19	activities	activity	NNS
kim	203	20	promise	promise	VBP
kim	203	21	a	a	DT
kim	203	22	higher	high	JJR
kim	203	23	level	level	NN
kim	203	24	of	of	IN
kim	203	25	personalization	personalization	NN
kim	203	26	.	.	.
kim	204	1	But	but	CC
kim	204	2	a	a	DT
kim	204	3	highly	highly	RB
kim	204	4	personalized	personalized	JJ
kim	204	5	information	information	NN
kim	204	6	environment	environment	NN
kim	204	7	often	often	RB
kim	204	8	traps	trap	VBZ
kim	204	9	people	people	NNS
kim	204	10	in	in	IN
kim	204	11	their	-PRON-	PRP$
kim	204	12	own	own	JJ
kim	204	13	so	so	RB
kim	204	14	-	-	HYPH
kim	204	15	called	call	VBN
kim	204	16	‘	'	``
kim	204	17	filter	filter	NN
kim	204	18	bubble	bubble	NN
kim	204	19	,	,	,
kim	204	20	’	'	''
kim	204	21	as	as	IN
kim	204	22	we	-PRON-	PRP
kim	204	23	have	have	VBP
kim	204	24	been	be	VBN
kim	204	25	increasingly	increasingly	RB
kim	204	26	seeing	see	VBG
kim	204	27	in	in	IN
kim	204	28	today	today	NN
kim	204	29	’s	’s	POS
kim	204	30	social	social	JJ
kim	204	31	media	medium	NNS
kim	204	32	channels	channel	NNS
kim	204	33	,	,	,
kim	204	34	news	news	NN
kim	204	35	websites	website	NNS
kim	204	36	,	,	,
kim	204	37	and	and	CC
kim	204	38	commercial	commercial	JJ
kim	204	39	search	search	NN
kim	204	40	engines	engine	NNS
kim	204	41	,	,	,
kim	204	42	where	where	WRB
kim	204	43	such	such	JJ
kim	204	44	personalization	personalization	NN
kim	204	45	is	be	VBZ
kim	204	46	provided	provide	VBN
kim	204	47	by	by	IN
kim	204	48	machine	machine	NN
kim	204	49	learning	learning	NN
kim	204	50	and	and	CC
kim	204	51	deep	deep	JJ
kim	204	52	learning	learning	NN
kim	204	53	.	.	.
kim	205	1	[	[	-LRB-
kim	205	2	footnoteRef:20	footnoteRef:20	NNP
kim	205	3	]	]	-RRB-
kim	205	4	Sophisticated	Sophisticated	NNP
kim	205	5	AI	AI	NNP
kim	205	6	algorithms	algorithm	NNS
kim	205	7	are	be	VBP
kim	205	8	already	already	RB
kim	205	9	curating	curate	VBG
kim	205	10	and	and	CC
kim	205	11	pushing	push	VBG
kim	205	12	information	information	NN
kim	205	13	feeds	feed	VBZ
kim	205	14	based	base	VBN
kim	205	15	upon	upon	IN
kim	205	16	the	the	DT
kim	205	17	person	person	NN
kim	205	18	’s	’s	POS
kim	205	19	past	past	JJ
kim	205	20	search	search	NN
kim	205	21	and	and	CC
kim	205	22	click	click	VB
kim	205	23	behavior	behavior	NN
kim	205	24	.	.	.
kim	206	1	The	the	DT
kim	206	2	result	result	NN
kim	206	3	is	be	VBZ
kim	206	4	that	that	IN
kim	206	5	information	information	NN
kim	206	6	seekers	seeker	NNS
kim	206	7	are	be	VBP
kim	206	8	provided	provide	VBN
kim	206	9	with	with	IN
kim	206	10	information	information	NN
kim	206	11	that	that	WDT
kim	206	12	conforms	conform	VBZ
kim	206	13	and	and	CC
kim	206	14	reinforces	reinforce	VBZ
kim	206	15	their	-PRON-	PRP$
kim	206	16	existing	exist	VBG
kim	206	17	beliefs	belief	NNS
kim	206	18	and	and	CC
kim	206	19	interests	interest	NNS
kim	206	20	.	.	.
kim	207	1	Views	view	NNS
kim	207	2	that	that	WDT
kim	207	3	are	be	VBP
kim	207	4	novel	novel	JJ
kim	207	5	or	or	CC
kim	207	6	contrast	contrast	NN
kim	207	7	with	with	IN
kim	207	8	their	-PRON-	PRP$
kim	207	9	existing	exist	VBG
kim	207	10	beliefs	belief	NNS
kim	207	11	are	be	VBP
kim	207	12	increasingly	increasingly	RB
kim	207	13	suppressed	suppress	VBN
kim	207	14	and	and	CC
kim	207	15	become	become	VBN
kim	207	16	invisible	invisible	JJ
kim	207	17	without	without	IN
kim	207	18	them	-PRON-	PRP
kim	207	19	even	even	RB
kim	207	20	realizing	realize	VBG
kim	207	21	.	.	.
kim	208	1	[	[	-LRB-
kim	208	2	20	20	CD
kim	208	3	:	:	:
kim	208	4	See	see	VB
kim	208	5	Pariser	Pariser	NNP
kim	208	6	2012	2012	CD
kim	208	7	.	.	.
kim	208	8	]	]	-RRB-
kim	209	1	Such	such	JJ
kim	209	2	lack	lack	NN
kim	209	3	of	of	IN
kim	209	4	exposure	exposure	NN
kim	209	5	to	to	IN
kim	209	6	opposing	oppose	VBG
kim	209	7	views	view	NNS
kim	209	8	leads	lead	VBZ
kim	209	9	information	information	NN
kim	209	10	users	user	NNS
kim	209	11	to	to	IN
kim	209	12	intellectual	intellectual	JJ
kim	209	13	isolation	isolation	NN
kim	209	14	and	and	CC
kim	209	15	even	even	RB
kim	209	16	bigotry	bigotry	NN
kim	209	17	.	.	.
kim	210	1	Highly	highly	RB
kim	210	2	personalized	personalized	JJ
kim	210	3	information	information	NN
kim	210	4	environments	environment	NNS
kim	210	5	powered	power	VBN
kim	210	6	by	by	IN
kim	210	7	AI	AI	NNP
kim	210	8	can	can	MD
kim	210	9	actively	actively	RB
kim	210	10	restrict	restrict	VB
kim	210	11	ways	way	NNS
kim	210	12	in	in	IN
kim	210	13	which	which	WDT
kim	210	14	people	people	NNS
kim	210	15	develop	develop	VBP
kim	210	16	balanced	balanced	JJ
kim	210	17	and	and	CC
kim	210	18	informed	informed	JJ
kim	210	19	opinions	opinion	NNS
kim	210	20	,	,	,
kim	210	21	thereby	thereby	RB
kim	210	22	intensifying	intensify	VBG
kim	210	23	and	and	CC
kim	210	24	perpetuating	perpetuate	VBG
kim	210	25	social	social	JJ
kim	210	26	discord	discord	NN
kim	210	27	and	and	CC
kim	210	28	disrupting	disrupt	VBG
kim	210	29	civic	civic	JJ
kim	210	30	discourse	discourse	NN
kim	210	31	.	.	.
kim	211	1	Such	such	JJ
kim	211	2	social	social	JJ
kim	211	3	discord	discord	NN
kim	211	4	and	and	CC
kim	211	5	disruption	disruption	NN
kim	211	6	in	in	IN
kim	211	7	civil	civil	JJ
kim	211	8	discourse	discourse	NN
kim	211	9	is	be	VBZ
kim	211	10	likely	likely	JJ
kim	211	11	to	to	TO
kim	211	12	disproportionately	disproportionately	RB
kim	211	13	affect	affect	VB
kim	211	14	those	those	DT
kim	211	15	with	with	IN
kim	211	16	least	least	JJS
kim	211	17	privileges	privilege	NNS
kim	211	18	.	.	.
kim	212	1	In	in	IN
kim	212	2	this	this	DT
kim	212	3	sense	sense	NN
kim	212	4	,	,	,
kim	212	5	intellectual	intellectual	JJ
kim	212	6	isolation	isolation	NN
kim	212	7	and	and	CC
kim	212	8	bigotry	bigotry	NN
kim	212	9	has	have	VBZ
kim	212	10	a	a	DT
kim	212	11	distinctly	distinctly	RB
kim	212	12	moral	moral	JJ
kim	212	13	impact	impact	NN
kim	212	14	on	on	IN
kim	212	15	society	society	NN
kim	212	16	.	.	.
kim	213	1	(	(	-LRB-
kim	213	2	ii	ii	NNP
kim	213	3	)	)	-RRB-
kim	213	4	Weakening	Weakening	NNP
kim	213	5	of	of	IN
kim	213	6	cognitive	cognitive	JJ
kim	213	7	agency	agency	NN
kim	213	8	and	and	CC
kim	213	9	autonomy	autonomy	NN
kim	213	10	We	-PRON-	PRP
kim	213	11	have	have	VBP
kim	213	12	seen	see	VBN
kim	213	13	earlier	early	RBR
kim	213	14	that	that	IN
kim	213	15	AI	AI	NNP
kim	213	16	-	-	HYPH
kim	213	17	powered	power	VBN
kim	213	18	digital	digital	JJ
kim	213	19	assistants	assistant	NNS
kim	213	20	are	be	VBP
kim	213	21	likely	likely	JJ
kim	213	22	to	to	TO
kim	213	23	mediate	mediate	VB
kim	213	24	people	people	NNS
kim	213	25	’s	’s	POS
kim	213	26	information	information	NN
kim	213	27	search	search	NN
kim	213	28	,	,	,
kim	213	29	discovery	discovery	NN
kim	213	30	,	,	,
kim	213	31	and	and	CC
kim	213	32	retrieval	retrieval	NN
kim	213	33	activities	activity	NNS
kim	213	34	in	in	IN
kim	213	35	the	the	DT
kim	213	36	near	near	JJ
kim	213	37	future	future	NN
kim	213	38	.	.	.
kim	214	1	As	as	IN
kim	214	2	those	those	DT
kim	214	3	digital	digital	JJ
kim	214	4	assistants	assistant	NNS
kim	214	5	become	become	VBP
kim	214	6	more	more	RBR
kim	214	7	capable	capable	JJ
kim	214	8	,	,	,
kim	214	9	they	-PRON-	PRP
kim	214	10	will	will	MD
kim	214	11	go	go	VB
kim	214	12	beyond	beyond	IN
kim	214	13	listing	list	VBG
kim	214	14	available	available	JJ
kim	214	15	information	information	NN
kim	214	16	.	.	.
kim	215	1	They	-PRON-	PRP
kim	215	2	will	will	MD
kim	215	3	further	further	RB
kim	215	4	choose	choose	VB
kim	215	5	what	what	WP
kim	215	6	they	-PRON-	PRP
kim	215	7	deem	deem	VBP
kim	215	8	to	to	TO
kim	215	9	be	be	VB
kim	215	10	most	most	RBS
kim	215	11	relevant	relevant	JJ
kim	215	12	to	to	IN
kim	215	13	their	-PRON-	PRP$
kim	215	14	users	user	NNS
kim	215	15	and	and	CC
kim	215	16	proceed	proceed	VB
kim	215	17	to	to	TO
kim	215	18	recommend	recommend	VB
kim	215	19	or	or	CC
kim	215	20	autonomously	autonomously	RB
kim	215	21	execute	execute	VB
kim	215	22	the	the	DT
kim	215	23	best	good	JJS
kim	215	24	course	course	NN
kim	215	25	of	of	IN
kim	215	26	action	action	NN
kim	215	27	.	.	.
kim	216	1	[	[	-LRB-
kim	216	2	footnoteRef:21	footnoteRef:21	NNP
kim	216	3	]	]	-RRB-
kim	216	4	Other	other	JJ
kim	216	5	AI	ai	NN
kim	216	6	-	-	HYPH
kim	216	7	driven	drive	VBN
kim	216	8	features	feature	NNS
kim	216	9	,	,	,
kim	216	10	such	such	JJ
kim	216	11	as	as	IN
kim	216	12	extracting	extract	VBG
kim	216	13	key	key	JJ
kim	216	14	information	information	NN
kim	216	15	or	or	CC
kim	216	16	generating	generate	VBG
kim	216	17	a	a	DT
kim	216	18	summary	summary	NN
kim	216	19	of	of	IN
kim	216	20	a	a	DT
kim	216	21	large	large	JJ
kim	216	22	amount	amount	NN
kim	216	23	of	of	IN
kim	216	24	information	information	NN
kim	216	25	,	,	,
kim	216	26	are	be	VBP
kim	216	27	also	also	RB
kim	216	28	likely	likely	JJ
kim	216	29	to	to	TO
kim	216	30	be	be	VB
kim	216	31	included	include	VBN
kim	216	32	in	in	IN
kim	216	33	future	future	JJ
kim	216	34	information	information	NN
kim	216	35	systems	system	NNS
kim	216	36	,	,	,
kim	216	37	and	and	CC
kim	216	38	they	-PRON-	PRP
kim	216	39	may	may	MD
kim	216	40	deliver	deliver	VB
kim	216	41	key	key	JJ
kim	216	42	information	information	NN
kim	216	43	or	or	CC
kim	216	44	summaries	summary	NNS
kim	216	45	even	even	RB
kim	216	46	before	before	IN
kim	216	47	the	the	DT
kim	216	48	request	request	NN
kim	216	49	is	be	VBZ
kim	216	50	made	make	VBN
kim	216	51	,	,	,
kim	216	52	based	base	VBN
kim	216	53	upon	upon	IN
kim	216	54	constant	constant	JJ
kim	216	55	monitoring	monitoring	NN
kim	216	56	of	of	IN
kim	216	57	the	the	DT
kim	216	58	user	user	NN
kim	216	59	’s	’s	POS
kim	216	60	activities	activity	NNS
kim	216	61	.	.	.
kim	217	1	[	[	-LRB-
kim	217	2	21	21	CD
kim	217	3	:	:	:
kim	217	4	Please	please	UH
kim	217	5	note	note	VB
kim	217	6	that	that	IN
kim	217	7	this	this	DT
kim	217	8	is	be	VBZ
kim	217	9	a	a	DT
kim	217	10	simplified	simplified	JJ
kim	217	11	scenario	scenario	NN
kim	217	12	.	.	.
kim	218	1	The	the	DT
kim	218	2	features	feature	NNS
kim	218	3	can	can	MD
kim	218	4	also	also	RB
kim	218	5	be	be	VB
kim	218	6	built	build	VBN
kim	218	7	in	in	IN
kim	218	8	the	the	DT
kim	218	9	information	information	NN
kim	218	10	system	system	NN
kim	218	11	itself	-PRON-	PRP
kim	218	12	rather	rather	RB
kim	218	13	than	than	IN
kim	218	14	the	the	DT
kim	218	15	agent	agent	NN
kim	218	16	.	.	.
kim	218	17	]	]	-RRB-
kim	219	1	In	in	IN
kim	219	2	such	such	PDT
kim	219	3	a	a	DT
kim	219	4	scenario	scenario	NN
kim	219	5	,	,	,
kim	219	6	an	an	DT
kim	219	7	information	information	NN
kim	219	8	seeker	seeker	NN
kim	219	9	’s	’s	POS
kim	219	10	cognitive	cognitive	JJ
kim	219	11	agency	agency	NN
kim	219	12	is	be	VBZ
kim	219	13	likely	likely	JJ
kim	219	14	to	to	TO
kim	219	15	suffer	suffer	VB
kim	219	16	a	a	DT
kim	219	17	loss	loss	NN
kim	219	18	.	.	.
kim	220	1	Crucial	crucial	JJ
kim	220	2	to	to	IN
kim	220	3	cognitive	cognitive	JJ
kim	220	4	agency	agency	NN
kim	220	5	is	be	VBZ
kim	220	6	the	the	DT
kim	220	7	mental	mental	JJ
kim	220	8	capacity	capacity	NN
kim	220	9	to	to	TO
kim	220	10	critically	critically	RB
kim	220	11	review	review	VB
kim	220	12	a	a	DT
kim	220	13	variety	variety	NN
kim	220	14	of	of	IN
kim	220	15	information	information	NN
kim	220	16	,	,	,
kim	220	17	judge	judge	VBP
kim	220	18	what	what	WP
kim	220	19	is	be	VBZ
kim	220	20	and	and	CC
kim	220	21	is	be	VBZ
kim	220	22	not	not	RB
kim	220	23	relevant	relevant	JJ
kim	220	24	,	,	,
kim	220	25	and	and	CC
kim	220	26	interpret	interpret	VB
kim	220	27	how	how	WRB
kim	220	28	they	-PRON-	PRP
kim	220	29	relate	relate	VBP
kim	220	30	to	to	IN
kim	220	31	other	other	JJ
kim	220	32	existing	exist	VBG
kim	220	33	beliefs	belief	NNS
kim	220	34	and	and	CC
kim	220	35	opinions	opinion	NNS
kim	220	36	.	.	.
kim	221	1	If	if	IN
kim	221	2	AI	AI	NNP
kim	221	3	assumes	assume	VBZ
kim	221	4	those	those	DT
kim	221	5	tasks	task	NNS
kim	221	6	,	,	,
kim	221	7	the	the	DT
kim	221	8	opportunities	opportunity	NNS
kim	221	9	for	for	IN
kim	221	10	information	information	NN
kim	221	11	seekers	seeker	NNS
kim	221	12	to	to	TO
kim	221	13	exercise	exercise	VB
kim	221	14	their	-PRON-	PRP$
kim	221	15	own	own	JJ
kim	221	16	cognitive	cognitive	JJ
kim	221	17	agency	agency	NN
kim	221	18	will	will	MD
kim	221	19	surely	surely	RB
kim	221	20	decrease	decrease	VB
kim	221	21	.	.	.
kim	222	1	Cognitive	cognitive	JJ
kim	222	2	deskilling	deskilling	NN
kim	222	3	and	and	CC
kim	222	4	the	the	DT
kim	222	5	subsequent	subsequent	JJ
kim	222	6	weakening	weakening	NN
kim	222	7	of	of	IN
kim	222	8	people	people	NNS
kim	222	9	’s	’s	POS
kim	222	10	agency	agency	NN
kim	222	11	in	in	IN
kim	222	12	the	the	DT
kim	222	13	AI	AI	NNP
kim	222	14	-powered	-powere	VBN
kim	222	15	automated	automate	VBN
kim	222	16	information	information	NN
kim	222	17	environment	environment	NN
kim	222	18	presents	present	VBZ
kim	222	19	an	an	DT
kim	222	20	ethical	ethical	JJ
kim	222	21	challenge	challenge	NN
kim	222	22	because	because	IN
kim	222	23	such	such	JJ
kim	222	24	agency	agency	NN
kim	222	25	is	be	VBZ
kim	222	26	necessary	necessary	JJ
kim	222	27	for	for	IN
kim	222	28	a	a	DT
kim	222	29	person	person	NN
kim	222	30	to	to	TO
kim	222	31	be	be	VB
kim	222	32	a	a	DT
kim	222	33	fully	fully	RB
kim	222	34	functioning	function	VBG
kim	222	35	moral	moral	JJ
kim	222	36	agent	agent	NN
kim	222	37	in	in	IN
kim	222	38	society	society	NN
kim	222	39	.	.	.
kim	223	1	[	[	-LRB-
kim	223	2	footnoteRef:22	footnoteref:22	NN
kim	223	3	]	]	-RRB-
kim	223	4	[	[	-LRB-
kim	223	5	22	22	CD
kim	223	6	:	:	:
kim	223	7	Outside	outside	IN
kim	223	8	of	of	IN
kim	223	9	the	the	DT
kim	223	10	automated	automated	JJ
kim	223	11	information	information	NN
kim	223	12	environment	environment	NN
kim	223	13	,	,	,
kim	223	14	AI	AI	NNP
kim	223	15	has	have	VBZ
kim	223	16	a	a	DT
kim	223	17	strong	strong	JJ
kim	223	18	potential	potential	NN
kim	223	19	to	to	TO
kim	223	20	engender	engender	VB
kim	223	21	moral	moral	JJ
kim	223	22	deskilling	deskilling	NN
kim	223	23	.	.	.
kim	224	1	Vallor	Vallor	NNP
kim	224	2	(	(	-LRB-
kim	224	3	2015	2015	CD
kim	224	4	)	)	-RRB-
kim	224	5	points	point	VBZ
kim	224	6	out	out	RP
kim	224	7	that	that	IN
kim	224	8	automated	automate	VBN
kim	224	9	weapons	weapon	NNS
kim	224	10	will	will	MD
kim	224	11	lead	lead	VB
kim	224	12	to	to	IN
kim	224	13	soldiers	soldier	NNS
kim	224	14	’	’	POS
kim	224	15	moral	moral	JJ
kim	224	16	deskilling	deskilling	NN
kim	224	17	in	in	IN
kim	224	18	the	the	DT
kim	224	19	use	use	NN
kim	224	20	of	of	IN
kim	224	21	military	military	JJ
kim	224	22	force	force	NN
kim	224	23	;	;	:
kim	224	24	new	new	JJ
kim	224	25	media	medium	NNS
kim	224	26	practices	practice	NNS
kim	224	27	of	of	IN
kim	224	28	multitasking	multitasking	NN
kim	224	29	may	may	MD
kim	224	30	result	result	VB
kim	224	31	in	in	IN
kim	224	32	deskilling	deskille	VBG
kim	224	33	in	in	IN
kim	224	34	moral	moral	JJ
kim	224	35	attention	attention	NN
kim	224	36	;	;	:
kim	224	37	and	and	CC
kim	224	38	social	social	JJ
kim	224	39	robots	robot	NNS
kim	224	40	can	can	MD
kim	224	41	cause	cause	VB
kim	224	42	moral	moral	JJ
kim	224	43	deskilling	deskilling	NN
kim	224	44	in	in	IN
kim	224	45	practices	practice	NNS
kim	224	46	of	of	IN
kim	224	47	human	human	JJ
kim	224	48	caregiving	caregiving	NN
kim	224	49	.	.	.
kim	224	50	]	]	-RRB-
kim	225	1	(	(	-LRB-
kim	225	2	iii	iii	NNP
kim	225	3	)	)	-RRB-
kim	225	4	Social	social	JJ
kim	225	5	impact	impact	NN
kim	225	6	of	of	IN
kim	225	7	scholarship	scholarship	NN
kim	225	8	and	and	CC
kim	225	9	research	research	NN
kim	225	10	from	from	IN
kim	225	11	flawed	flawed	JJ
kim	225	12	AI	AI	NNP
kim	225	13	algorithms	algorithm	NNS
kim	225	14	Previously	previously	RB
kim	225	15	,	,	,
kim	225	16	we	-PRON-	PRP
kim	225	17	have	have	VBP
kim	225	18	seen	see	VBN
kim	225	19	that	that	IN
kim	225	20	deep	deep	JJ
kim	225	21	learning	learning	NN
kim	225	22	applications	application	NNS
kim	225	23	are	be	VBP
kim	225	24	opaque	opaque	JJ
kim	225	25	to	to	IN
kim	225	26	human	human	JJ
kim	225	27	understanding	understanding	NN
kim	225	28	.	.	.
kim	226	1	This	this	DT
kim	226	2	lack	lack	NN
kim	226	3	of	of	IN
kim	226	4	transparency	transparency	NN
kim	226	5	and	and	CC
kim	226	6	explainability	explainability	NN
kim	226	7	raises	raise	VBZ
kim	226	8	the	the	DT
kim	226	9	question	question	NN
kim	226	10	of	of	IN
kim	226	11	whether	whether	IN
kim	226	12	it	-PRON-	PRP
kim	226	13	is	be	VBZ
kim	226	14	moral	moral	JJ
kim	226	15	to	to	TO
kim	226	16	rely	rely	VB
kim	226	17	on	on	IN
kim	226	18	AI	AI	NNP
kim	226	19	-	-	HYPH
kim	226	20	powered	power	VBN
kim	226	21	military	military	JJ
kim	226	22	robots	robot	NNS
kim	226	23	for	for	IN
kim	226	24	life	life	NN
kim	226	25	-	-	HYPH
kim	226	26	or	or	CC
kim	226	27	-	-	HYPH
kim	226	28	death	death	NN
kim	226	29	decisions	decision	NNS
kim	226	30	.	.	.
kim	227	1	Does	do	VBZ
kim	227	2	the	the	DT
kim	227	3	AI	ai	NN
kim	227	4	-	-	HYPH
kim	227	5	powered	power	VBN
kim	227	6	information	information	NN
kim	227	7	environment	environment	NN
kim	227	8	have	have	VBP
kim	227	9	a	a	DT
kim	227	10	similar	similar	JJ
kim	227	11	problem	problem	NN
kim	227	12	?	?	.
kim	228	1	As	as	IN
kim	228	2	explained	explain	VBN
kim	228	3	in	in	IN
kim	228	4	the	the	DT
kim	228	5	beginning	beginning	NN
kim	228	6	of	of	IN
kim	228	7	this	this	DT
kim	228	8	chapter	chapter	NN
kim	228	9	,	,	,
kim	228	10	machine	machine	NN
kim	228	11	learning	learning	NN
kim	228	12	applications	application	NNS
kim	228	13	base	base	VBP
kim	228	14	their	-PRON-	PRP$
kim	228	15	recommendations	recommendation	NNS
kim	228	16	and	and	CC
kim	228	17	predictions	prediction	NNS
kim	228	18	upon	upon	IN
kim	228	19	the	the	DT
kim	228	20	patterns	pattern	NNS
kim	228	21	in	in	IN
kim	228	22	past	past	JJ
kim	228	23	data	datum	NNS
kim	228	24	.	.	.
kim	229	1	Their	-PRON-	PRP$
kim	229	2	predictions	prediction	NNS
kim	229	3	and	and	CC
kim	229	4	recommendations	recommendation	NNS
kim	229	5	are	be	VBP
kim	229	6	in	in	IN
kim	229	7	this	this	DT
kim	229	8	sense	sense	NN
kim	229	9	inherently	inherently	RB
kim	229	10	conservative	conservative	JJ
kim	229	11	and	and	CC
kim	229	12	become	become	VBP
kim	229	13	outdated	outdated	JJ
kim	229	14	when	when	WRB
kim	229	15	they	-PRON-	PRP
kim	229	16	fail	fail	VBP
kim	229	17	to	to	TO
kim	229	18	reflect	reflect	VB
kim	229	19	the	the	DT
kim	229	20	new	new	JJ
kim	229	21	social	social	JJ
kim	229	22	views	view	NNS
kim	229	23	and	and	CC
kim	229	24	material	material	NN
kim	229	25	conditions	condition	NNS
kim	229	26	that	that	WDT
kim	229	27	no	no	RB
kim	229	28	longer	long	RBR
kim	229	29	fit	fit	VBP
kim	229	30	the	the	DT
kim	229	31	past	past	JJ
kim	229	32	patterns	pattern	NNS
kim	229	33	.	.	.
kim	230	1	Furthermore	furthermore	RB
kim	230	2	,	,	,
kim	230	3	each	each	DT
kim	230	4	data	data	NN
kim	230	5	set	set	NN
kim	230	6	is	be	VBZ
kim	230	7	a	a	DT
kim	230	8	social	social	JJ
kim	230	9	construct	construct	NN
kim	230	10	that	that	WDT
kim	230	11	reflects	reflect	VBZ
kim	230	12	particular	particular	JJ
kim	230	13	values	value	NNS
kim	230	14	and	and	CC
kim	230	15	choices	choice	NNS
kim	230	16	such	such	JJ
kim	230	17	as	as	IN
kim	230	18	who	who	WP
kim	230	19	decided	decide	VBD
kim	230	20	to	to	TO
kim	230	21	collect	collect	VB
kim	230	22	the	the	DT
kim	230	23	data	datum	NNS
kim	230	24	and	and	CC
kim	230	25	for	for	IN
kim	230	26	what	what	WDT
kim	230	27	purpose	purpose	NN
kim	230	28	;	;	:
kim	230	29	who	who	WP
kim	230	30	labeled	label	VBD
kim	230	31	data	datum	NNS
kim	230	32	;	;	:
kim	230	33	what	what	WDT
kim	230	34	criteria	criterion	NNS
kim	230	35	or	or	CC
kim	230	36	beliefs	belief	NNS
kim	230	37	guided	guide	VBD
kim	230	38	such	such	JJ
kim	230	39	labeling	labeling	NN
kim	230	40	;	;	:
kim	230	41	what	what	WP
kim	230	42	taxonomies	taxonomy	NNS
kim	230	43	were	be	VBD
kim	230	44	used	use	VBN
kim	230	45	and	and	CC
kim	230	46	why	why	WRB
kim	230	47	(	(	-LRB-
kim	230	48	Davis	Davis	NNP
kim	230	49	2020	2020	CD
kim	230	50	)	)	-RRB-
kim	230	51	.	.	.
kim	231	1	No	no	DT
kim	231	2	data	data	NN
kim	231	3	set	set	NN
kim	231	4	can	can	MD
kim	231	5	capture	capture	VB
kim	231	6	all	all	PDT
kim	231	7	the	the	DT
kim	231	8	variables	variable	NNS
kim	231	9	and	and	CC
kim	231	10	elements	element	NNS
kim	231	11	of	of	IN
kim	231	12	the	the	DT
kim	231	13	phenomenon	phenomenon	NN
kim	231	14	that	that	WDT
kim	231	15	it	-PRON-	PRP
kim	231	16	describes	describe	VBZ
kim	231	17	.	.	.
kim	232	1	Furthermore	furthermore	RB
kim	232	2	,	,	,
kim	232	3	data	data	NN
kim	232	4	sets	set	NNS
kim	232	5	used	use	VBN
kim	232	6	for	for	IN
kim	232	7	training	train	VBG
kim	232	8	machine	machine	NN
kim	232	9	learning	learning	NN
kim	232	10	and	and	CC
kim	232	11	deep	deep	JJ
kim	232	12	learning	learning	NN
kim	232	13	algorithms	algorithm	NNS
kim	232	14	may	may	MD
kim	232	15	not	not	RB
kim	232	16	be	be	VB
kim	232	17	representational	representational	JJ
kim	232	18	samples	sample	NNS
kim	232	19	for	for	IN
kim	232	20	all	all	DT
kim	232	21	relevant	relevant	JJ
kim	232	22	subgroups	subgroup	NNS
kim	232	23	.	.	.
kim	233	1	In	in	IN
kim	233	2	such	such	PDT
kim	233	3	a	a	DT
kim	233	4	case	case	NN
kim	233	5	,	,	,
kim	233	6	an	an	DT
kim	233	7	algorithm	algorithm	NN
kim	233	8	will	will	MD
kim	233	9	not	not	RB
kim	233	10	be	be	VB
kim	233	11	able	able	JJ
kim	233	12	to	to	TO
kim	233	13	perform	perform	VB
kim	233	14	equally	equally	RB
kim	233	15	well	well	RB
kim	233	16	in	in	IN
kim	233	17	all	all	DT
kim	233	18	subgroups	subgroup	NNS
kim	233	19	.	.	.
kim	234	1	Creating	create	VBG
kim	234	2	a	a	DT
kim	234	3	large	large	JJ
kim	234	4	data	data	NN
kim	234	5	set	set	NN
kim	234	6	is	be	VBZ
kim	234	7	also	also	RB
kim	234	8	costly	costly	JJ
kim	234	9	.	.	.
kim	235	1	Consequently	consequently	RB
kim	235	2	,	,	,
kim	235	3	developers	developer	NNS
kim	235	4	often	often	RB
kim	235	5	simply	simply	RB
kim	235	6	take	take	VBP
kim	235	7	data	data	NN
kim	235	8	sets	set	NNS
kim	235	9	that	that	WDT
kim	235	10	are	be	VBP
kim	235	11	available	available	JJ
kim	235	12	to	to	IN
kim	235	13	them	-PRON-	PRP
kim	235	14	.	.	.
kim	236	1	Those	those	DT
kim	236	2	data	data	NN
kim	236	3	sets	set	NNS
kim	236	4	are	be	VBP
kim	236	5	likely	likely	JJ
kim	236	6	to	to	TO
kim	236	7	come	come	VB
kim	236	8	with	with	IN
kim	236	9	inherent	inherent	JJ
kim	236	10	limitations	limitation	NNS
kim	236	11	such	such	JJ
kim	236	12	as	as	IN
kim	236	13	omissions	omission	NNS
kim	236	14	,	,	,
kim	236	15	inaccuracies	inaccuracy	NNS
kim	236	16	,	,	,
kim	236	17	errors	error	NNS
kim	236	18	,	,	,
kim	236	19	and	and	CC
kim	236	20	hidden	hide	VBN
kim	236	21	biases	bias	NNS
kim	236	22	.	.	.
kim	237	1	AI	AI	NNP
kim	237	2	algorithms	algorithm	NNS
kim	237	3	trained	train	VBD
kim	237	4	with	with	IN
kim	237	5	these	these	DT
kim	237	6	flawed	flawed	JJ
kim	237	7	data	data	NN
kim	237	8	sets	set	NNS
kim	237	9	can	can	MD
kim	237	10	fail	fail	VB
kim	237	11	unexpectedly	unexpectedly	RB
kim	237	12	,	,	,
kim	237	13	revealing	reveal	VBG
kim	237	14	those	those	DT
kim	237	15	limitations	limitation	NNS
kim	237	16	in	in	IN
kim	237	17	the	the	DT
kim	237	18	data	data	NN
kim	237	19	sets	set	NNS
kim	237	20	.	.	.
kim	238	1	For	for	IN
kim	238	2	example	example	NN
kim	238	3	,	,	,
kim	238	4	it	-PRON-	PRP
kim	238	5	has	have	VBZ
kim	238	6	been	be	VBN
kim	238	7	reported	report	VBN
kim	238	8	that	that	IN
kim	238	9	the	the	DT
kim	238	10	success	success	NN
kim	238	11	rate	rate	NN
kim	238	12	of	of	IN
kim	238	13	a	a	DT
kim	238	14	facial	facial	JJ
kim	238	15	recognition	recognition	NN
kim	238	16	algorithm	algorithm	NN
kim	238	17	plunges	plunge	NNS
kim	238	18	from	from	IN
kim	238	19	99	99	CD
kim	238	20	%	%	NN
kim	238	21	to	to	TO
kim	238	22	35	35	CD
kim	238	23	%	%	NN
kim	238	24	when	when	WRB
kim	238	25	the	the	DT
kim	238	26	group	group	NN
kim	238	27	of	of	IN
kim	238	28	subjects	subject	NNS
kim	238	29	changes	change	NNS
kim	238	30	from	from	IN
kim	238	31	white	white	JJ
kim	238	32	men	man	NNS
kim	238	33	to	to	IN
kim	238	34	dark	dark	JJ
kim	238	35	-	-	HYPH
kim	238	36	skinned	skinned	JJ
kim	238	37	women	woman	NNS
kim	238	38	because	because	IN
kim	238	39	it	-PRON-	PRP
kim	238	40	was	be	VBD
kim	238	41	trained	train	VBN
kim	238	42	mostly	mostly	RB
kim	238	43	with	with	IN
kim	238	44	the	the	DT
kim	238	45	photographs	photograph	NNS
kim	238	46	of	of	IN
kim	238	47	white	white	JJ
kim	238	48	men	man	NNS
kim	238	49	(	(	-LRB-
kim	238	50	Lohr	Lohr	NNP
kim	238	51	2018	2018	CD
kim	238	52	)	)	-RRB-
kim	238	53	.	.	.
kim	239	1	Adopting	adopt	VBG
kim	239	2	such	such	PDT
kim	239	3	a	a	DT
kim	239	4	faulty	faulty	JJ
kim	239	5	algorithm	algorithm	NN
kim	239	6	for	for	IN
kim	239	7	any	any	DT
kim	239	8	real	real	JJ
kim	239	9	-	-	HYPH
kim	239	10	life	life	NN
kim	239	11	use	use	NN
kim	239	12	at	at	IN
kim	239	13	a	a	DT
kim	239	14	large	large	JJ
kim	239	15	scale	scale	NN
kim	239	16	would	would	MD
kim	239	17	be	be	VB
kim	239	18	entirely	entirely	RB
kim	239	19	unethical	unethical	JJ
kim	239	20	.	.	.
kim	240	1	For	for	IN
kim	240	2	the	the	DT
kim	240	3	context	context	NN
kim	240	4	of	of	IN
kim	240	5	libraries	library	NNS
kim	240	6	,	,	,
kim	240	7	imagine	imagine	VB
kim	240	8	using	use	VBG
kim	240	9	such	such	PDT
kim	240	10	a	a	DT
kim	240	11	face	face	NN
kim	240	12	-	-	HYPH
kim	240	13	recognition	recognition	NN
kim	240	14	algorithm	algorithm	NN
kim	240	15	to	to	TO
kim	240	16	generate	generate	VB
kim	240	17	metadata	metadata	NN
kim	240	18	for	for	IN
kim	240	19	digitized	digitized	JJ
kim	240	20	historical	historical	JJ
kim	240	21	photographs	photograph	NNS
kim	240	22	,	,	,
kim	240	23	or	or	CC
kim	240	24	a	a	DT
kim	240	25	similarly	similarly	RB
kim	240	26	flawed	flawed	JJ
kim	240	27	audio	audio	JJ
kim	240	28	transcription	transcription	NN
kim	240	29	algorithm	algorithm	NNP
kim	240	30	to	to	TO
kim	240	31	transcribe	transcribe	VB
kim	240	32	archival	archival	NNP
kim	240	33	audio	audio	NNP
kim	240	34	recordings	recording	NNS
kim	240	35	.	.	.
kim	241	1	Just	just	RB
kim	241	2	like	like	IN
kim	241	3	those	those	DT
kim	241	4	faulty	faulty	JJ
kim	241	5	algorithms	algorithm	NNS
kim	241	6	,	,	,
kim	241	7	an	an	DT
kim	241	8	AI	AI	NNP
kim	241	9	-	-	HYPH
kim	241	10	powered	power	VBN
kim	241	11	automated	automated	JJ
kim	241	12	information	information	NN
kim	241	13	environment	environment	NN
kim	241	14	can	can	MD
kim	241	15	provide	provide	VB
kim	241	16	information	information	NN
kim	241	17	,	,	,
kim	241	18	recommendations	recommendation	NNS
kim	241	19	,	,	,
kim	241	20	and	and	CC
kim	241	21	predictions	prediction	NNS
kim	241	22	affected	affect	VBN
kim	241	23	by	by	IN
kim	241	24	similar	similar	JJ
kim	241	25	limitations	limitation	NNS
kim	241	26	existing	exist	VBG
kim	241	27	in	in	IN
kim	241	28	many	many	JJ
kim	241	29	data	datum	NNS
kim	241	30	sets	set	NNS
kim	241	31	.	.	.
kim	242	1	The	the	DT
kim	242	2	more	more	RBR
kim	242	3	seamless	seamless	NN
kim	242	4	such	such	PDT
kim	242	5	an	an	DT
kim	242	6	information	information	NN
kim	242	7	environment	environment	NN
kim	242	8	is	be	VBZ
kim	242	9	,	,	,
kim	242	10	the	the	DT
kim	242	11	more	more	RBR
kim	242	12	invisible	invisible	JJ
kim	242	13	those	those	DT
kim	242	14	limitations	limitation	NNS
kim	242	15	become	become	VBP
kim	242	16	.	.	.
kim	243	1	Automated	automate	VBN
kim	243	2	information	information	NN
kim	243	3	systems	system	NNS
kim	243	4	from	from	IN
kim	243	5	libraries	library	NNS
kim	243	6	may	may	MD
kim	243	7	not	not	RB
kim	243	8	be	be	VB
kim	243	9	involved	involve	VBN
kim	243	10	in	in	IN
kim	243	11	decisions	decision	NNS
kim	243	12	that	that	WDT
kim	243	13	have	have	VBP
kim	243	14	a	a	DT
kim	243	15	direct	direct	JJ
kim	243	16	and	and	CC
kim	243	17	immediate	immediate	JJ
kim	243	18	impact	impact	NN
kim	243	19	on	on	IN
kim	243	20	people	people	NNS
kim	243	21	’s	’s	POS
kim	243	22	lives	life	NNS
kim	243	23	,	,	,
kim	243	24	such	such	JJ
kim	243	25	as	as	IN
kim	243	26	setting	set	VBG
kim	243	27	a	a	DT
kim	243	28	bail	bail	NN
kim	243	29	amount	amount	NN
kim	243	30	or	or	CC
kim	243	31	determining	determine	VBG
kim	243	32	the	the	DT
kim	243	33	Medicaid	Medicaid	NNP
kim	243	34	payment	payment	NN
kim	243	35	to	to	TO
kim	243	36	be	be	VB
kim	243	37	paid,[footnoteRef:23	paid,[footnoteref:23	CD
kim	243	38	]	]	-RRB-
kim	243	39	but	but	CC
kim	243	40	automated	automate	VBD
kim	243	41	information	information	NN
kim	243	42	systems	system	NNS
kim	243	43	that	that	WDT
kim	243	44	are	be	VBP
kim	243	45	widely	widely	RB
kim	243	46	adopted	adopt	VBN
kim	243	47	and	and	CC
kim	243	48	used	use	VBN
kim	243	49	for	for	IN
kim	243	50	research	research	NN
kim	243	51	and	and	CC
kim	243	52	scholarship	scholarship	NN
kim	243	53	will	will	MD
kim	243	54	impact	impact	VB
kim	243	55	real	real	JJ
kim	243	56	-	-	HYPH
kim	243	57	life	life	NN
kim	243	58	policies	policy	NNS
kim	243	59	and	and	CC
kim	243	60	regulations	regulation	NNS
kim	243	61	in	in	IN
kim	243	62	areas	area	NNS
kim	243	63	such	such	JJ
kim	243	64	as	as	IN
kim	243	65	healthcare	healthcare	NN
kim	243	66	and	and	CC
kim	243	67	the	the	DT
kim	243	68	economy	economy	NN
kim	243	69	.	.	.
kim	244	1	Undiscovered	undiscovered	JJ
kim	244	2	flaws	flaw	NNS
kim	244	3	will	will	MD
kim	244	4	undermine	undermine	VB
kim	244	5	the	the	DT
kim	244	6	validity	validity	NN
kim	244	7	of	of	IN
kim	244	8	the	the	DT
kim	244	9	scholarly	scholarly	JJ
kim	244	10	output	output	NN
kim	244	11	that	that	WDT
kim	244	12	utilized	utilize	VBD
kim	244	13	those	those	DT
kim	244	14	automated	automate	VBN
kim	244	15	information	information	NN
kim	244	16	systems	system	NNS
kim	244	17	and	and	CC
kim	244	18	can	can	MD
kim	244	19	further	far	RBR
kim	244	20	inflict	inflict	VB
kim	244	21	serious	serious	JJ
kim	244	22	harm	harm	NN
kim	244	23	on	on	IN
kim	244	24	certain	certain	JJ
kim	244	25	groups	group	NNS
kim	244	26	of	of	IN
kim	244	27	people	people	NNS
kim	244	28	through	through	IN
kim	244	29	those	those	DT
kim	244	30	policies	policy	NNS
kim	244	31	and	and	CC
kim	244	32	regulations	regulation	NNS
kim	244	33	.	.	.
kim	245	1	[	[	-LRB-
kim	245	2	23	23	CD
kim	245	3	:	:	:
kim	245	4	See	see	VB
kim	245	5	Tashea	Tashea	NNP
kim	245	6	2017	2017	CD
kim	245	7	and	and	CC
kim	245	8	Stanley	Stanley	NNP
kim	245	9	2017	2017	CD
kim	245	10	.	.	.
kim	245	11	]	]	-RRB-
kim	246	1	7	7	LS
kim	246	2	.	.	.
kim	247	1	Moral	Moral	NNP
kim	247	2	Intelligence	Intelligence	NNP
kim	247	3	and	and	CC
kim	247	4	Rethinking	rethink	VBG
kim	247	5	the	the	DT
kim	247	6	Role	Role	NNP
kim	247	7	of	of	IN
kim	247	8	AI	AI	NNP
kim	247	9	In	in	IN
kim	247	10	this	this	DT
kim	247	11	chapter	chapter	NN
kim	247	12	,	,	,
kim	247	13	I	-PRON-	PRP
kim	247	14	discussed	discuss	VBD
kim	247	15	four	four	CD
kim	247	16	significant	significant	JJ
kim	247	17	ethical	ethical	JJ
kim	247	18	challenges	challenge	NNS
kim	247	19	that	that	IN
kim	247	20	automating	automate	VBG
kim	247	21	decisions	decision	NNS
kim	247	22	and	and	CC
kim	247	23	actions	action	NNS
kim	247	24	with	with	IN
kim	247	25	AI	AI	NNP
kim	247	26	presents	present	NNS
kim	247	27	:	:	:
kim	247	28	(	(	-LRB-
kim	247	29	a	a	DT
kim	247	30	)	)	-RRB-
kim	247	31	moral	moral	JJ
kim	247	32	desensitization	desensitization	NN
kim	247	33	;	;	,
kim	247	34	(	(	-LRB-
kim	247	35	b	b	NN
kim	247	36	)	)	-RRB-
kim	247	37	unintended	unintended	JJ
kim	247	38	outcomes	outcome	NNS
kim	247	39	;	;	,
kim	247	40	(	(	-LRB-
kim	247	41	c	c	NN
kim	247	42	)	)	-RRB-
kim	247	43	surrender	surrender	NN
kim	247	44	of	of	IN
kim	247	45	moral	moral	JJ
kim	247	46	agency	agency	NN
kim	247	47	;	;	:
kim	247	48	(	(	-LRB-
kim	247	49	d	d	NN
kim	247	50	)	)	-RRB-
kim	247	51	opacity	opacity	NN
kim	247	52	in	in	IN
kim	247	53	decision	decision	NN
kim	247	54	-	-	HYPH
kim	247	55	making	making	NN
kim	247	56	.	.	.
kim	248	1	[	[	-LRB-
kim	248	2	footnoteRef:24	footnoteRef:24	NNP
kim	248	3	]	]	-RRB-
kim	248	4	I	-PRON-	PRP
kim	248	5	also	also	RB
kim	248	6	examined	examine	VBD
kim	248	7	somewhat	somewhat	RB
kim	248	8	different	different	JJ
kim	248	9	but	but	CC
kim	248	10	equally	equally	RB
kim	248	11	significant	significant	JJ
kim	248	12	ethical	ethical	JJ
kim	248	13	challenges	challenge	NNS
kim	248	14	in	in	IN
kim	248	15	relation	relation	NN
kim	248	16	to	to	IN
kim	248	17	the	the	DT
kim	248	18	AI	AI	NNP
kim	248	19	-	-	HYPH
kim	248	20	powered	power	VBN
kim	248	21	automated	automated	JJ
kim	248	22	information	information	NN
kim	248	23	environment	environment	NN
kim	248	24	,	,	,
kim	248	25	which	which	WDT
kim	248	26	is	be	VBZ
kim	248	27	likely	likely	JJ
kim	248	28	to	to	TO
kim	248	29	surround	surround	VB
kim	248	30	us	-PRON-	PRP
kim	248	31	in	in	IN
kim	248	32	the	the	DT
kim	248	33	future	future	NN
kim	248	34	:	:	:
kim	248	35	(	(	-LRB-
kim	248	36	i	i	NN
kim	248	37	)	)	-RRB-
kim	248	38	intellectual	intellectual	JJ
kim	248	39	isolation	isolation	NN
kim	248	40	and	and	CC
kim	248	41	bigotry	bigotry	NN
kim	248	42	hampering	hamper	VBG
kim	248	43	civic	civic	JJ
kim	248	44	discourse	discourse	NN
kim	248	45	;	;	:
kim	248	46	(	(	-LRB-
kim	248	47	ii	ii	NN
kim	248	48	)	)	-RRB-
kim	248	49	weakening	weaken	VBG
kim	248	50	of	of	IN
kim	248	51	cognitive	cognitive	JJ
kim	248	52	agency	agency	NN
kim	248	53	and	and	CC
kim	248	54	autonomy	autonomy	NN
kim	248	55	;	;	,
kim	248	56	(	(	-LRB-
kim	248	57	iii	iii	NNP
kim	248	58	)	)	-RRB-
kim	248	59	social	social	JJ
kim	248	60	impact	impact	NN
kim	248	61	of	of	IN
kim	248	62	scholarship	scholarship	NN
kim	248	63	and	and	CC
kim	248	64	research	research	NN
kim	248	65	based	base	VBN
kim	248	66	upon	upon	IN
kim	248	67	flawed	flawed	JJ
kim	248	68	AI	AI	NNP
kim	248	69	algorithms	algorithm	NNS
kim	248	70	.	.	.
kim	249	1	[	[	-LRB-
kim	249	2	24	24	CD
kim	249	3	:	:	:
kim	249	4	This	this	DT
kim	249	5	is	be	VBZ
kim	249	6	by	by	IN
kim	249	7	no	no	DT
kim	249	8	means	means	NN
kim	249	9	an	an	DT
kim	249	10	exhaustive	exhaustive	JJ
kim	249	11	list	list	NN
kim	249	12	.	.	.
kim	250	1	User	user	NN
kim	250	2	privacy	privacy	NN
kim	250	3	and	and	CC
kim	250	4	potential	potential	JJ
kim	250	5	surveillance	surveillance	NN
kim	250	6	are	be	VBP
kim	250	7	examples	example	NNS
kim	250	8	of	of	IN
kim	250	9	other	other	JJ
kim	250	10	important	important	JJ
kim	250	11	ethical	ethical	JJ
kim	250	12	challenges	challenge	NNS
kim	250	13	,	,	,
kim	250	14	which	which	WDT
kim	250	15	I	-PRON-	PRP
kim	250	16	do	do	VBP
kim	250	17	not	not	RB
kim	250	18	discuss	discuss	VB
kim	250	19	here	here	RB
kim	250	20	.	.	.
kim	250	21	]	]	-RRB-
kim	251	1	The	the	DT
kim	251	2	challenges	challenge	NNS
kim	251	3	related	relate	VBN
kim	251	4	to	to	IN
kim	251	5	the	the	DT
kim	251	6	AI	AI	NNP
kim	251	7	-	-	HYPH
kim	251	8	powered	power	VBN
kim	251	9	automated	automated	JJ
kim	251	10	information	information	NN
kim	251	11	environment	environment	NN
kim	251	12	are	be	VBP
kim	251	13	relevant	relevant	JJ
kim	251	14	to	to	IN
kim	251	15	libraries	library	NNS
kim	251	16	,	,	,
kim	251	17	which	which	WDT
kim	251	18	will	will	MD
kim	251	19	soon	soon	RB
kim	251	20	be	be	VB
kim	251	21	acquiring	acquire	VBG
kim	251	22	,	,	,
kim	251	23	building	build	VBG
kim	251	24	,	,	,
kim	251	25	customizing	customizing	NN
kim	251	26	,	,	,
kim	251	27	and	and	CC
kim	251	28	implementing	implement	VBG
kim	251	29	many	many	JJ
kim	251	30	personalized	personalize	VBN
kim	251	31	and	and	CC
kim	251	32	automated	automate	VBD
kim	251	33	information	information	NN
kim	251	34	systems	system	NNS
kim	251	35	for	for	IN
kim	251	36	their	-PRON-	PRP$
kim	251	37	users	user	NNS
kim	251	38	.	.	.
kim	252	1	At	at	IN
kim	252	2	present	present	NN
kim	252	3	,	,	,
kim	252	4	libraries	library	NNS
kim	252	5	are	be	VBP
kim	252	6	at	at	IN
kim	252	7	an	an	DT
kim	252	8	early	early	JJ
kim	252	9	stage	stage	NN
kim	252	10	in	in	IN
kim	252	11	developing	develop	VBG
kim	252	12	AI	ai	NN
kim	252	13	applications	application	NNS
kim	252	14	and	and	CC
kim	252	15	applying	apply	VBG
kim	252	16	machine	machine	NN
kim	252	17	learning	learning	NN
kim	252	18	and	and	CC
kim	252	19	deep	deep	JJ
kim	252	20	learning	learning	NN
kim	252	21	techniques	technique	NNS
kim	252	22	to	to	TO
kim	252	23	improve	improve	VB
kim	252	24	library	library	JJ
kim	252	25	services	service	NNS
kim	252	26	,	,	,
kim	252	27	systems	system	NNS
kim	252	28	,	,	,
kim	252	29	and	and	CC
kim	252	30	operations	operation	NNS
kim	252	31	.	.	.
kim	253	1	But	but	CC
kim	253	2	the	the	DT
kim	253	3	general	general	JJ
kim	253	4	issues	issue	NNS
kim	253	5	of	of	IN
kim	253	6	hidden	hide	VBN
kim	253	7	biases	bias	NNS
kim	253	8	and	and	CC
kim	253	9	the	the	DT
kim	253	10	lack	lack	NN
kim	253	11	of	of	IN
kim	253	12	explainability	explainability	NN
kim	253	13	in	in	IN
kim	253	14	machine	machine	NN
kim	253	15	learning	learning	NN
kim	253	16	and	and	CC
kim	253	17	deep	deep	JJ
kim	253	18	learning	learning	NN
kim	253	19	are	be	VBP
kim	253	20	quickly	quickly	RB
kim	253	21	gaining	gain	VBG
kim	253	22	awareness	awareness	NN
kim	253	23	in	in	IN
kim	253	24	the	the	DT
kim	253	25	library	library	NN
kim	253	26	community	community	NN
kim	253	27	.	.	.
kim	254	1	As	as	IN
kim	254	2	we	-PRON-	PRP
kim	254	3	have	have	VBP
kim	254	4	seen	see	VBN
kim	254	5	in	in	IN
kim	254	6	the	the	DT
kim	254	7	trolley	trolley	NN
kim	254	8	problem	problem	NN
kim	254	9	,	,	,
kim	254	10	whether	whether	IN
kim	254	11	a	a	DT
kim	254	12	certain	certain	JJ
kim	254	13	action	action	NN
kim	254	14	is	be	VBZ
kim	254	15	moral	moral	JJ
kim	254	16	or	or	CC
kim	254	17	not	not	RB
kim	254	18	is	be	VBZ
kim	254	19	a	a	DT
kim	254	20	line	line	NN
kim	254	21	that	that	WDT
kim	254	22	can	can	MD
kim	254	23	not	not	RB
kim	254	24	be	be	VB
kim	254	25	drawn	draw	VBN
kim	254	26	with	with	IN
kim	254	27	absolute	absolute	JJ
kim	254	28	clarity	clarity	NN
kim	254	29	.	.	.
kim	255	1	It	-PRON-	PRP
kim	255	2	is	be	VBZ
kim	255	3	entirely	entirely	RB
kim	255	4	possible	possible	JJ
kim	255	5	for	for	IN
kim	255	6	fully	fully	RB
kim	255	7	-	-	HYPH
kim	255	8	functioning	function	VBG
kim	255	9	moral	moral	JJ
kim	255	10	agents	agent	NNS
kim	255	11	to	to	TO
kim	255	12	make	make	VB
kim	255	13	different	different	JJ
kim	255	14	judgements	judgement	NNS
kim	255	15	in	in	IN
kim	255	16	a	a	DT
kim	255	17	particular	particular	JJ
kim	255	18	situation	situation	NN
kim	255	19	.	.	.
kim	256	1	Additionally	additionally	RB
kim	256	2	,	,	,
kim	256	3	there	there	EX
kim	256	4	is	be	VBZ
kim	256	5	also	also	RB
kim	256	6	the	the	DT
kim	256	7	matter	matter	NN
kim	256	8	of	of	IN
kim	256	9	morality	morality	NN
kim	256	10	that	that	IN
kim	256	11	our	-PRON-	PRP$
kim	256	12	tools	tool	NNS
kim	256	13	and	and	CC
kim	256	14	systems	system	NNS
kim	256	15	display	display	NN
kim	256	16	.	.	.
kim	257	1	This	this	DT
kim	257	2	is	be	VBZ
kim	257	3	called	call	VBN
kim	257	4	machine	machine	NN
kim	257	5	morality	morality	NN
kim	257	6	in	in	IN
kim	257	7	relation	relation	NN
kim	257	8	to	to	IN
kim	257	9	AI	AI	NNP
kim	257	10	systems	system	NNS
kim	257	11	,	,	,
kim	257	12	since	since	IN
kim	257	13	AI	AI	NNP
kim	257	14	systems	system	NNS
kim	257	15	are	be	VBP
kim	257	16	intelligent	intelligent	JJ
kim	257	17	machines	machine	NNS
kim	257	18	.	.	.
kim	258	1	Wallach	Wallach	NNP
kim	258	2	and	and	CC
kim	258	3	Allen	Allen	NNP
kim	258	4	(	(	-LRB-
kim	258	5	2009	2009	CD
kim	258	6	)	)	-RRB-
kim	258	7	argue	argue	VBP
kim	258	8	that	that	IN
kim	258	9	there	there	EX
kim	258	10	are	be	VBP
kim	258	11	three	three	CD
kim	258	12	distinct	distinct	JJ
kim	258	13	levels	level	NNS
kim	258	14	of	of	IN
kim	258	15	machine	machine	NN
kim	258	16	morality	morality	NN
kim	258	17	:	:	:
kim	258	18	operational	operational	JJ
kim	258	19	morality	morality	NN
kim	258	20	,	,	,
kim	258	21	functional	functional	JJ
kim	258	22	morality	morality	NN
kim	258	23	,	,	,
kim	258	24	and	and	CC
kim	258	25	full	full	JJ
kim	258	26	moral	moral	JJ
kim	258	27	agency	agency	NN
kim	258	28	(	(	-LRB-
kim	258	29	26	26	CD
kim	258	30	)	)	-RRB-
kim	258	31	.	.	.
kim	259	1	Operational	operational	JJ
kim	259	2	morality	morality	NN
kim	259	3	is	be	VBZ
kim	259	4	found	find	VBN
kim	259	5	in	in	IN
kim	259	6	systems	system	NNS
kim	259	7	that	that	WDT
kim	259	8	are	be	VBP
kim	259	9	low	low	JJ
kim	259	10	in	in	IN
kim	259	11	both	both	CC
kim	259	12	autonomy	autonomy	NN
kim	259	13	and	and	CC
kim	259	14	ethical	ethical	JJ
kim	259	15	sensitivity	sensitivity	NN
kim	259	16	.	.	.
kim	260	1	At	at	IN
kim	260	2	this	this	DT
kim	260	3	level	level	NN
kim	260	4	of	of	IN
kim	260	5	machine	machine	NN
kim	260	6	morality	morality	NN
kim	260	7	,	,	,
kim	260	8	a	a	DT
kim	260	9	machine	machine	NN
kim	260	10	or	or	CC
kim	260	11	a	a	DT
kim	260	12	tool	tool	NN
kim	260	13	is	be	VBZ
kim	260	14	given	give	VBN
kim	260	15	a	a	DT
kim	260	16	mechanism	mechanism	NN
kim	260	17	that	that	WDT
kim	260	18	prevents	prevent	VBZ
kim	260	19	its	-PRON-	PRP$
kim	260	20	immoral	immoral	JJ
kim	260	21	use	use	NN
kim	260	22	,	,	,
kim	260	23	but	but	CC
kim	260	24	the	the	DT
kim	260	25	mechanism	mechanism	NN
kim	260	26	is	be	VBZ
kim	260	27	within	within	IN
kim	260	28	the	the	DT
kim	260	29	full	full	JJ
kim	260	30	control	control	NN
kim	260	31	of	of	IN
kim	260	32	the	the	DT
kim	260	33	machine	machine	NN
kim	260	34	or	or	CC
kim	260	35	the	the	DT
kim	260	36	tool	tool	NN
kim	260	37	’s	’s	POS
kim	260	38	designers	designer	NNS
kim	260	39	and	and	CC
kim	260	40	users	user	NNS
kim	260	41	.	.	.
kim	261	1	Such	such	JJ
kim	261	2	operational	operational	JJ
kim	261	3	morality	morality	NN
kim	261	4	exists	exist	VBZ
kim	261	5	in	in	IN
kim	261	6	a	a	DT
kim	261	7	gun	gun	NN
kim	261	8	with	with	IN
kim	261	9	a	a	DT
kim	261	10	childproof	childproof	JJ
kim	261	11	safety	safety	NN
kim	261	12	mechanism	mechanism	NN
kim	261	13	,	,	,
kim	261	14	for	for	IN
kim	261	15	example	example	NN
kim	261	16	.	.	.
kim	262	1	A	a	DT
kim	262	2	gun	gun	NN
kim	262	3	with	with	IN
kim	262	4	a	a	DT
kim	262	5	safety	safety	NN
kim	262	6	mechanism	mechanism	NN
kim	262	7	is	be	VBZ
kim	262	8	hardly	hardly	RB
kim	262	9	autonomous	autonomous	JJ
kim	262	10	nor	nor	CC
kim	262	11	sensitive	sensitive	JJ
kim	262	12	to	to	IN
kim	262	13	any	any	DT
kim	262	14	ethical	ethical	JJ
kim	262	15	concerns	concern	NNS
kim	262	16	related	relate	VBN
kim	262	17	to	to	IN
kim	262	18	its	-PRON-	PRP$
kim	262	19	use	use	NN
kim	262	20	.	.	.
kim	263	1	By	by	IN
kim	263	2	contrast	contrast	NN
kim	263	3	,	,	,
kim	263	4	machines	machine	NNS
kim	263	5	with	with	IN
kim	263	6	functional	functional	JJ
kim	263	7	morality	morality	NN
kim	263	8	do	do	VBP
kim	263	9	possess	possess	VB
kim	263	10	a	a	DT
kim	263	11	certain	certain	JJ
kim	263	12	level	level	NN
kim	263	13	of	of	IN
kim	263	14	autonomy	autonomy	NN
kim	263	15	and	and	CC
kim	263	16	ethical	ethical	JJ
kim	263	17	sensitivity	sensitivity	NN
kim	263	18	.	.	.
kim	264	1	They	-PRON-	PRP
kim	264	2	may	may	MD
kim	264	3	include	include	VB
kim	264	4	an	an	DT
kim	264	5	AI	AI	NNP
kim	264	6	system	system	NN
kim	264	7	with	with	IN
kim	264	8	significant	significant	JJ
kim	264	9	autonomy	autonomy	NN
kim	264	10	but	but	CC
kim	264	11	little	little	JJ
kim	264	12	ethical	ethical	JJ
kim	264	13	sensitivity	sensitivity	NN
kim	264	14	or	or	CC
kim	264	15	an	an	DT
kim	264	16	AI	AI	NNP
kim	264	17	system	system	NN
kim	264	18	with	with	IN
kim	264	19	little	little	JJ
kim	264	20	autonomy	autonomy	NN
kim	264	21	but	but	CC
kim	264	22	high	high	JJ
kim	264	23	ethical	ethical	JJ
kim	264	24	sensitivity	sensitivity	NN
kim	264	25	.	.	.
kim	265	1	An	an	DT
kim	265	2	autonomous	autonomous	JJ
kim	265	3	drone	drone	NN
kim	265	4	would	would	MD
kim	265	5	fall	fall	VB
kim	265	6	under	under	IN
kim	265	7	the	the	DT
kim	265	8	former	former	JJ
kim	265	9	category	category	NN
kim	265	10	,	,	,
kim	265	11	while	while	IN
kim	265	12	MedEthEx	medethex	NN
kim	265	13	,	,	,
kim	265	14	an	an	DT
kim	265	15	ethical	ethical	JJ
kim	265	16	decision	decision	NN
kim	265	17	-	-	HYPH
kim	265	18	support	support	NN
kim	265	19	AI	AI	NNP
kim	265	20	recommendation	recommendation	NN
kim	265	21	system	system	NN
kim	265	22	for	for	IN
kim	265	23	clinicians	clinician	NNS
kim	265	24	,	,	,
kim	265	25	would	would	MD
kim	265	26	be	be	VB
kim	265	27	of	of	IN
kim	265	28	the	the	DT
kim	265	29	latter	latter	JJ
kim	265	30	.	.	.
kim	266	1	Lastly	lastly	RB
kim	266	2	,	,	,
kim	266	3	systems	system	NNS
kim	266	4	with	with	IN
kim	266	5	high	high	JJ
kim	266	6	autonomy	autonomy	NN
kim	266	7	and	and	CC
kim	266	8	high	high	JJ
kim	266	9	ethical	ethical	JJ
kim	266	10	sensitivity	sensitivity	NN
kim	266	11	can	can	MD
kim	266	12	be	be	VB
kim	266	13	regarded	regard	VBN
kim	266	14	as	as	IN
kim	266	15	having	have	VBG
kim	266	16	full	full	JJ
kim	266	17	moral	moral	JJ
kim	266	18	agency	agency	NN
kim	266	19	,	,	,
kim	266	20	as	as	RB
kim	266	21	much	much	RB
kim	266	22	as	as	IN
kim	266	23	humans	human	NNS
kim	266	24	do	do	VBP
kim	266	25	.	.	.
kim	267	1	This	this	DT
kim	267	2	means	mean	VBZ
kim	267	3	that	that	IN
kim	267	4	those	those	DT
kim	267	5	systems	system	NNS
kim	267	6	would	would	MD
kim	267	7	have	have	VB
kim	267	8	a	a	DT
kim	267	9	mental	mental	JJ
kim	267	10	representation	representation	NN
kim	267	11	of	of	IN
kim	267	12	values	value	NNS
kim	267	13	and	and	CC
kim	267	14	the	the	DT
kim	267	15	capacity	capacity	NN
kim	267	16	for	for	IN
kim	267	17	moral	moral	JJ
kim	267	18	reasoning	reasoning	NN
kim	267	19	.	.	.
kim	268	1	Such	such	JJ
kim	268	2	machines	machine	NNS
kim	268	3	can	can	MD
kim	268	4	be	be	VB
kim	268	5	held	hold	VBN
kim	268	6	morally	morally	RB
kim	268	7	responsible	responsible	JJ
kim	268	8	for	for	IN
kim	268	9	their	-PRON-	PRP$
kim	268	10	actions	action	NNS
kim	268	11	.	.	.
kim	269	1	We	-PRON-	PRP
kim	269	2	do	do	VBP
kim	269	3	not	not	RB
kim	269	4	know	know	VB
kim	269	5	whether	whether	IN
kim	269	6	AI	AI	NNP
kim	269	7	will	will	MD
kim	269	8	be	be	VB
kim	269	9	able	able	JJ
kim	269	10	to	to	TO
kim	269	11	produce	produce	VB
kim	269	12	such	such	PDT
kim	269	13	a	a	DT
kim	269	14	machine	machine	NN
kim	269	15	with	with	IN
kim	269	16	full	full	JJ
kim	269	17	moral	moral	JJ
kim	269	18	agency	agency	NN
kim	269	19	.	.	.
kim	270	1	If	if	IN
kim	270	2	the	the	DT
kim	270	3	current	current	JJ
kim	270	4	direction	direction	NN
kim	270	5	to	to	TO
kim	270	6	automate	automate	VB
kim	270	7	more	more	JJR
kim	270	8	and	and	CC
kim	270	9	more	more	JJR
kim	270	10	human	human	JJ
kim	270	11	tasks	task	NNS
kim	270	12	for	for	IN
kim	270	13	cost	cost	NN
kim	270	14	savings	saving	NNS
kim	270	15	and	and	CC
kim	270	16	efficiency	efficiency	NN
kim	270	17	at	at	IN
kim	270	18	scale	scale	NN
kim	270	19	continues	continue	VBZ
kim	270	20	,	,	,
kim	270	21	however	however	RB
kim	270	22	,	,	,
kim	270	23	most	most	JJS
kim	270	24	of	of	IN
kim	270	25	the	the	DT
kim	270	26	more	more	RBR
kim	270	27	sophisticated	sophisticated	JJ
kim	270	28	AI	AI	NNP
kim	270	29	applications	application	NNS
kim	270	30	to	to	TO
kim	270	31	come	come	VB
kim	270	32	will	will	MD
kim	270	33	be	be	VB
kim	270	34	of	of	IN
kim	270	35	the	the	DT
kim	270	36	kind	kind	NN
kim	270	37	with	with	IN
kim	270	38	functional	functional	JJ
kim	270	39	morality	morality	NN
kim	270	40	,	,	,
kim	270	41	particularly	particularly	RB
kim	270	42	the	the	DT
kim	270	43	kind	kind	NN
kim	270	44	that	that	WDT
kim	270	45	combines	combine	VBZ
kim	270	46	a	a	DT
kim	270	47	relatively	relatively	RB
kim	270	48	high	high	JJ
kim	270	49	level	level	NN
kim	270	50	of	of	IN
kim	270	51	autonomy	autonomy	NN
kim	270	52	and	and	CC
kim	270	53	a	a	DT
kim	270	54	lower	low	JJR
kim	270	55	level	level	NN
kim	270	56	of	of	IN
kim	270	57	ethical	ethical	JJ
kim	270	58	sensitivity	sensitivity	NN
kim	270	59	.	.	.
kim	271	1	In	in	IN
kim	271	2	the	the	DT
kim	271	3	beginning	beginning	NN
kim	271	4	of	of	IN
kim	271	5	this	this	DT
kim	271	6	chapter	chapter	NN
kim	271	7	,	,	,
kim	271	8	I	-PRON-	PRP
kim	271	9	mentioned	mention	VBD
kim	271	10	that	that	IN
kim	271	11	the	the	DT
kim	271	12	goal	goal	NN
kim	271	13	of	of	IN
kim	271	14	AI	AI	NNP
kim	271	15	is	be	VBZ
kim	271	16	to	to	TO
kim	271	17	create	create	VB
kim	271	18	an	an	DT
kim	271	19	artificial	artificial	JJ
kim	271	20	system	system	NN
kim	271	21	–	–	:
kim	271	22	whether	whether	IN
kim	271	23	it	-PRON-	PRP
kim	271	24	be	be	VB
kim	271	25	a	a	DT
kim	271	26	piece	piece	NN
kim	271	27	of	of	IN
kim	271	28	software	software	NN
kim	271	29	or	or	CC
kim	271	30	a	a	DT
kim	271	31	machine	machine	NN
kim	271	32	with	with	IN
kim	271	33	a	a	DT
kim	271	34	physical	physical	JJ
kim	271	35	body	body	NN
kim	271	36	–	–	:
kim	271	37	that	that	DT
kim	271	38	is	be	VBZ
kim	271	39	as	as	RB
kim	271	40	intelligent	intelligent	JJ
kim	271	41	as	as	IN
kim	271	42	a	a	DT
kim	271	43	human	human	NN
kim	271	44	in	in	IN
kim	271	45	its	-PRON-	PRP$
kim	271	46	performance	performance	NN
kim	271	47	,	,	,
kim	271	48	either	either	CC
kim	271	49	broadly	broadly	RB
kim	271	50	in	in	IN
kim	271	51	all	all	DT
kim	271	52	areas	area	NNS
kim	271	53	of	of	IN
kim	271	54	human	human	JJ
kim	271	55	activities	activity	NNS
kim	271	56	or	or	CC
kim	271	57	narrowly	narrowly	RB
kim	271	58	in	in	IN
kim	271	59	a	a	DT
kim	271	60	specific	specific	JJ
kim	271	61	activity	activity	NN
kim	271	62	,	,	,
kim	271	63	such	such	JJ
kim	271	64	as	as	IN
kim	271	65	playing	play	VBG
kim	271	66	chess	chess	NN
kim	271	67	or	or	CC
kim	271	68	driving	driving	NN
kim	271	69	.	.	.
kim	272	1	But	but	CC
kim	272	2	what	what	WP
kim	272	3	does	do	VBZ
kim	272	4	“	"	``
kim	272	5	as	as	RB
kim	272	6	intelligent	intelligent	JJ
kim	272	7	as	as	IN
kim	272	8	a	a	DT
kim	272	9	human	human	NN
kim	272	10	”	"	''
kim	272	11	exactly	exactly	RB
kim	272	12	mean	mean	VB
kim	272	13	?	?	.
kim	273	1	If	if	IN
kim	273	2	morality	morality	NN
kim	273	3	is	be	VBZ
kim	273	4	an	an	DT
kim	273	5	integral	integral	JJ
kim	273	6	component	component	NN
kim	273	7	of	of	IN
kim	273	8	human	human	JJ
kim	273	9	-	-	HYPH
kim	273	10	level	level	NN
kim	273	11	intelligence	intelligence	NN
kim	273	12	,	,	,
kim	273	13	AI	AI	NNP
kim	273	14	research	research	NN
kim	273	15	needs	need	VBZ
kim	273	16	to	to	TO
kim	273	17	pay	pay	VB
kim	273	18	more	more	JJR
kim	273	19	attention	attention	NN
kim	273	20	to	to	IN
kim	273	21	intelligence	intelligence	NN
kim	273	22	,	,	,
kim	273	23	not	not	RB
kim	273	24	just	just	RB
kim	273	25	in	in	IN
kim	273	26	accomplishing	accomplish	VBG
kim	273	27	a	a	DT
kim	273	28	goal	goal	NN
kim	273	29	but	but	CC
kim	273	30	also	also	RB
kim	273	31	in	in	IN
kim	273	32	doing	do	VBG
kim	273	33	so	so	RB
kim	273	34	ethically	ethically	RB
kim	273	35	.	.	.
kim	274	1	[	[	-LRB-
kim	274	2	footnoteRef:25	footnoteRef:25	NNP
kim	274	3	]	]	-RRB-
kim	274	4	In	in	IN
kim	274	5	that	that	DT
kim	274	6	light	light	NN
kim	274	7	,	,	,
kim	274	8	it	-PRON-	PRP
kim	274	9	is	be	VBZ
kim	274	10	meaningful	meaningful	JJ
kim	274	11	to	to	TO
kim	274	12	ask	ask	VB
kim	274	13	what	what	WDT
kim	274	14	level	level	NN
kim	274	15	of	of	IN
kim	274	16	autonomy	autonomy	NN
kim	274	17	and	and	CC
kim	274	18	ethical	ethical	JJ
kim	274	19	sensitivity	sensitivity	NN
kim	274	20	a	a	DT
kim	274	21	given	give	VBN
kim	274	22	AI	ai	NN
kim	274	23	system	system	NN
kim	274	24	is	be	VBZ
kim	274	25	equipped	equip	VBN
kim	274	26	with	with	IN
kim	274	27	,	,	,
kim	274	28	and	and	CC
kim	274	29	what	what	WDT
kim	274	30	level	level	NN
kim	274	31	of	of	IN
kim	274	32	machine	machine	NN
kim	274	33	morality	morality	NN
kim	274	34	is	be	VBZ
kim	274	35	appropriate	appropriate	JJ
kim	274	36	for	for	IN
kim	274	37	its	-PRON-	PRP$
kim	274	38	purpose	purpose	NN
kim	274	39	.	.	.
kim	275	1	[	[	-LRB-
kim	275	2	25	25	CD
kim	275	3	:	:	:
kim	275	4	Here	here	RB
kim	275	5	,	,	,
kim	275	6	I	-PRON-	PRP
kim	275	7	regard	regard	VBP
kim	275	8	intelligence	intelligence	NN
kim	275	9	as	as	IN
kim	275	10	the	the	DT
kim	275	11	ability	ability	NN
kim	275	12	to	to	TO
kim	275	13	accomplish	accomplish	VB
kim	275	14	complex	complex	JJ
kim	275	15	goals	goal	NNS
kim	275	16	following	follow	VBG
kim	275	17	Tegmark	Tegmark	NNP
kim	275	18	2017	2017	CD
kim	275	19	.	.	.
kim	276	1	For	for	IN
kim	276	2	more	more	JJR
kim	276	3	discussion	discussion	NN
kim	276	4	on	on	IN
kim	276	5	intelligence	intelligence	NN
kim	276	6	and	and	CC
kim	276	7	goals	goal	NNS
kim	276	8	,	,	,
kim	276	9	see	see	VB
kim	276	10	Chapter	chapter	NN
kim	276	11	2	2	CD
kim	276	12	and	and	CC
kim	276	13	Chapter	Chapter	NNP
kim	276	14	7	7	CD
kim	276	15	.	.	.
kim	276	16	]	]	-RRB-
kim	277	1	In	in	IN
kim	277	2	designing	design	VBG
kim	277	3	an	an	DT
kim	277	4	AI	AI	NNP
kim	277	5	system	system	NN
kim	277	6	,	,	,
kim	277	7	it	-PRON-	PRP
kim	277	8	would	would	MD
kim	277	9	be	be	VB
kim	277	10	also	also	RB
kim	277	11	helpful	helpful	JJ
kim	277	12	to	to	TO
kim	277	13	determine	determine	VB
kim	277	14	what	what	WDT
kim	277	15	level	level	NN
kim	277	16	of	of	IN
kim	277	17	autonomy	autonomy	NN
kim	277	18	and	and	CC
kim	277	19	ethical	ethical	JJ
kim	277	20	sensitivity	sensitivity	NN
kim	277	21	would	would	MD
kim	277	22	be	be	VB
kim	277	23	best	well	RBS
kim	277	24	suited	suit	VBN
kim	277	25	for	for	IN
kim	277	26	the	the	DT
kim	277	27	purpose	purpose	NN
kim	277	28	of	of	IN
kim	277	29	the	the	DT
kim	277	30	system	system	NN
kim	277	31	and	and	CC
kim	277	32	whether	whether	IN
kim	277	33	it	-PRON-	PRP
kim	277	34	is	be	VBZ
kim	277	35	feasible	feasible	JJ
kim	277	36	to	to	TO
kim	277	37	provide	provide	VB
kim	277	38	that	that	DT
kim	277	39	level	level	NN
kim	277	40	of	of	IN
kim	277	41	machine	machine	NN
kim	277	42	morality	morality	NN
kim	277	43	for	for	IN
kim	277	44	the	the	DT
kim	277	45	system	system	NN
kim	277	46	in	in	IN
kim	277	47	question	question	NN
kim	277	48	.	.	.
kim	278	1	The	the	DT
kim	278	2	narrower	narrow	JJR
kim	278	3	the	the	DT
kim	278	4	function	function	NN
kim	278	5	or	or	CC
kim	278	6	the	the	DT
kim	278	7	domain	domain	NN
kim	278	8	of	of	IN
kim	278	9	an	an	DT
kim	278	10	AI	AI	NNP
kim	278	11	system	system	NN
kim	278	12	is	be	VBZ
kim	278	13	,	,	,
kim	278	14	the	the	DT
kim	278	15	easier	easy	JJR
kim	278	16	it	-PRON-	PRP
kim	278	17	will	will	MD
kim	278	18	be	be	VB
kim	278	19	to	to	TO
kim	278	20	equip	equip	VB
kim	278	21	it	-PRON-	PRP
kim	278	22	with	with	IN
kim	278	23	an	an	DT
kim	278	24	appropriate	appropriate	JJ
kim	278	25	level	level	NN
kim	278	26	of	of	IN
kim	278	27	autonomy	autonomy	NN
kim	278	28	and	and	CC
kim	278	29	ethical	ethical	JJ
kim	278	30	sensitivity	sensitivity	NN
kim	278	31	.	.	.
kim	279	1	In	in	IN
kim	279	2	evaluating	evaluate	VBG
kim	279	3	and	and	CC
kim	279	4	designing	design	VBG
kim	279	5	an	an	DT
kim	279	6	AI	AI	NNP
kim	279	7	system	system	NN
kim	279	8	,	,	,
kim	279	9	it	-PRON-	PRP
kim	279	10	will	will	MD
kim	279	11	be	be	VB
kim	279	12	important	important	JJ
kim	279	13	to	to	TO
kim	279	14	test	test	VB
kim	279	15	the	the	DT
kim	279	16	actual	actual	JJ
kim	279	17	outcome	outcome	NN
kim	279	18	against	against	IN
kim	279	19	the	the	DT
kim	279	20	anticipated	anticipated	JJ
kim	279	21	outcome	outcome	NN
kim	279	22	in	in	IN
kim	279	23	different	different	JJ
kim	279	24	types	type	NNS
kim	279	25	of	of	IN
kim	279	26	cases	case	NNS
kim	279	27	in	in	IN
kim	279	28	order	order	NN
kim	279	29	to	to	TO
kim	279	30	identify	identify	VB
kim	279	31	potential	potential	JJ
kim	279	32	problems	problem	NNS
kim	279	33	.	.	.
kim	280	1	System	system	NN
kim	280	2	-	-	HYPH
kim	280	3	wide	wide	JJ
kim	280	4	audits	audits	NN
kim	280	5	to	to	TO
kim	280	6	detect	detect	VB
kim	280	7	well	well	RB
kim	280	8	-	-	HYPH
kim	280	9	known	know	VBN
kim	280	10	biases	bias	NNS
kim	280	11	,	,	,
kim	280	12	such	such	JJ
kim	280	13	as	as	IN
kim	280	14	gender	gender	NN
kim	280	15	discrimination	discrimination	NN
kim	280	16	or	or	CC
kim	280	17	racism	racism	NN
kim	280	18	can	can	MD
kim	280	19	serve	serve	VB
kim	280	20	as	as	IN
kim	280	21	an	an	DT
kim	280	22	effective	effective	JJ
kim	280	23	strategy	strategy	NN
kim	280	24	.	.	.
kim	281	1	[	[	-LRB-
kim	281	2	footnoteRef:26	footnoteRef:26	NNP
kim	281	3	]	]	-RRB-
kim	281	4	Other	other	JJ
kim	281	5	undetected	undetected	JJ
kim	281	6	problems	problem	NNS
kim	281	7	may	may	MD
kim	281	8	surface	surface	VB
kim	281	9	only	only	RB
kim	281	10	after	after	IN
kim	281	11	the	the	DT
kim	281	12	AI	AI	NNP
kim	281	13	system	system	NN
kim	281	14	is	be	VBZ
kim	281	15	deployed	deploy	VBN
kim	281	16	.	.	.
kim	282	1	Having	have	VBG
kim	282	2	a	a	DT
kim	282	3	mechanism	mechanism	NN
kim	282	4	to	to	TO
kim	282	5	continually	continually	RB
kim	282	6	test	test	VB
kim	282	7	an	an	DT
kim	282	8	AI	AI	NNP
kim	282	9	algorithm	algorithm	NN
kim	282	10	to	to	TO
kim	282	11	identify	identify	VB
kim	282	12	those	those	DT
kim	282	13	unnoticed	unnoticed	JJ
kim	282	14	problems	problem	NNS
kim	282	15	and	and	CC
kim	282	16	feeding	feed	VBG
kim	282	17	the	the	DT
kim	282	18	test	test	NN
kim	282	19	result	result	VB
kim	282	20	back	back	RB
kim	282	21	into	into	IN
kim	282	22	the	the	DT
kim	282	23	algorithm	algorithm	NN
kim	282	24	for	for	IN
kim	282	25	retraining	retraining	NN
kim	282	26	will	will	MD
kim	282	27	be	be	VB
kim	282	28	another	another	DT
kim	282	29	way	way	NN
kim	282	30	to	to	TO
kim	282	31	deal	deal	VB
kim	282	32	with	with	IN
kim	282	33	algorithmic	algorithmic	JJ
kim	282	34	biases	bias	NNS
kim	282	35	.	.	.
kim	283	1	Those	those	DT
kim	283	2	who	who	WP
kim	283	3	build	build	VBP
kim	283	4	AI	AI	NNP
kim	283	5	systems	system	NNS
kim	283	6	will	will	MD
kim	283	7	also	also	RB
kim	283	8	benefit	benefit	VB
kim	283	9	from	from	IN
kim	283	10	consulting	consult	VBG
kim	283	11	existing	exist	VBG
kim	283	12	principles	principle	NNS
kim	283	13	and	and	CC
kim	283	14	guidelines	guideline	NNS
kim	283	15	such	such	JJ
kim	283	16	as	as	IN
kim	283	17	FAT	FAT	NNP
kim	283	18	/	/	SYM
kim	283	19	ML	ML	NNP
kim	283	20	’s	’s	POS
kim	283	21	“	"	``
kim	283	22	Principles	principle	NNS
kim	283	23	for	for	IN
kim	283	24	Accountable	Accountable	NNP
kim	283	25	Algorithms	Algorithms	NNP
kim	283	26	and	and	CC
kim	283	27	a	a	DT
kim	283	28	Social	Social	NNP
kim	283	29	Impact	Impact	NNP
kim	283	30	Statement	Statement	NNP
kim	283	31	for	for	IN
kim	283	32	Algorithms	Algorithms	NNP
kim	283	33	.	.	.
kim	283	34	”[footnoteRef:27	”[footnoteRef:27	''
kim	283	35	]	]	-RRB-
kim	283	36	[	[	-LRB-
kim	283	37	26	26	CD
kim	283	38	:	:	:
kim	283	39	These	these	DT
kim	283	40	audits	audit	NNS
kim	283	41	are	be	VBP
kim	283	42	far	far	RB
kim	283	43	from	from	IN
kim	283	44	foolproof	foolproof	JJ
kim	283	45	,	,	,
kim	283	46	but	but	CC
kim	283	47	the	the	DT
kim	283	48	detection	detection	NN
kim	283	49	of	of	IN
kim	283	50	hidden	hidden	JJ
kim	283	51	biases	bias	NNS
kim	283	52	will	will	MD
kim	283	53	be	be	VB
kim	283	54	crucial	crucial	JJ
kim	283	55	in	in	IN
kim	283	56	making	make	VBG
kim	283	57	AI	AI	NNP
kim	283	58	algorithms	algorithm	NNS
kim	283	59	more	more	RBR
kim	283	60	accountable	accountable	JJ
kim	283	61	and	and	CC
kim	283	62	their	-PRON-	PRP$
kim	283	63	decisions	decision	NNS
kim	283	64	more	more	RBR
kim	283	65	ethical	ethical	JJ
kim	283	66	.	.	.
kim	284	1	A	a	DT
kim	284	2	debiasing	debiasing	JJ
kim	284	3	algorithm	algorithm	NN
kim	284	4	can	can	MD
kim	284	5	also	also	RB
kim	284	6	be	be	VB
kim	284	7	used	use	VBN
kim	284	8	during	during	IN
kim	284	9	the	the	DT
kim	284	10	training	training	NN
kim	284	11	stage	stage	NN
kim	284	12	of	of	IN
kim	284	13	an	an	DT
kim	284	14	AI	AI	NNP
kim	284	15	algorithm	algorithm	NN
kim	284	16	to	to	TO
kim	284	17	reduce	reduce	VB
kim	284	18	hidden	hide	VBN
kim	284	19	biases	bias	NNS
kim	284	20	in	in	IN
kim	284	21	training	training	NN
kim	284	22	data	datum	NNS
kim	284	23	.	.	.
kim	285	1	See	see	VB
kim	285	2	Amini	Amini	NNP
kim	285	3	et	et	NNP
kim	285	4	al	al	NNP
kim	285	5	.	.	.
kim	286	1	2019	2019	CD
kim	286	2	,	,	,
kim	286	3	Knight	Knight	NNP
kim	286	4	2019b	2019b	CD
kim	286	5	,	,	,
kim	286	6	and	and	CC
kim	286	7	Courtland	Courtland	NNP
kim	286	8	2018	2018	CD
kim	286	9	.	.	.
kim	286	10	]	]	-RRB-
kim	287	1	[	[	-LRB-
kim	287	2	27	27	CD
kim	287	3	:	:	:
kim	287	4	See	see	VB
kim	287	5	https://www.fatml.org/resources/principles-for-accountable-algorithms	https://www.fatml.org/resources/principles-for-accountable-algorithms	NNP
kim	287	6	.	.	.
kim	288	1	Other	other	JJ
kim	288	2	principles	principle	NNS
kim	288	3	and	and	CC
kim	288	4	guidelines	guideline	NNS
kim	288	5	include	include	VBP
kim	288	6	“	"	``
kim	288	7	Ethics	Ethics	NNP
kim	288	8	Guidelines	Guidelines	NNPS
kim	288	9	for	for	IN
kim	288	10	Trustworthy	Trustworthy	NNP
kim	288	11	AI	AI	NNP
kim	288	12	”	"	''
kim	288	13	(	(	-LRB-
kim	288	14	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
kim	288	15	)	)	-RRB-
kim	288	16	and	and	CC
kim	288	17	“	"	``
kim	288	18	Algorithmic	Algorithmic	NNP
kim	288	19	Impact	Impact	NNP
kim	288	20	Assessments	assessment	NNS
kim	288	21	:	:	:
kim	288	22	A	a	DT
kim	288	23	Practical	Practical	NNP
kim	288	24	Framework	Framework	NNP
kim	288	25	For	for	IN
kim	288	26	Public	Public	NNP
kim	288	27	Agency	Agency	NNP
kim	288	28	Accountability	Accountability	NNP
kim	288	29	”	"	''
kim	288	30	(	(	-LRB-
kim	288	31	https://ainowinstitute.org/aiareport2018.pdf	https://ainowinstitute.org/aiareport2018.pdf	NNP
kim	288	32	)	)	-RRB-
kim	288	33	.	.	.
kim	288	34	]	]	-RRB-
kim	289	1	We	-PRON-	PRP
kim	289	2	may	may	MD
kim	289	3	also	also	RB
kim	289	4	want	want	VB
kim	289	5	to	to	TO
kim	289	6	rethink	rethink	VB
kim	289	7	how	how	WRB
kim	289	8	and	and	CC
kim	289	9	where	where	WRB
kim	289	10	we	-PRON-	PRP
kim	289	11	apply	apply	VBP
kim	289	12	AI	AI	NNP
kim	289	13	.	.	.
kim	290	1	We	-PRON-	PRP
kim	290	2	and	and	CC
kim	290	3	our	-PRON-	PRP$
kim	290	4	society	society	NN
kim	290	5	do	do	VBP
kim	290	6	not	not	RB
kim	290	7	have	have	VB
kim	290	8	to	to	TO
kim	290	9	use	use	VB
kim	290	10	AI	AI	NNP
kim	290	11	to	to	TO
kim	290	12	equip	equip	VB
kim	290	13	all	all	PDT
kim	290	14	our	-PRON-	PRP$
kim	290	15	systems	system	NNS
kim	290	16	and	and	CC
kim	290	17	machines	machine	NNS
kim	290	18	with	with	IN
kim	290	19	human-	human-	NN
kim	290	20	or	or	CC
kim	290	21	superhuman	superhuman	JJ
kim	290	22	-	-	HYPH
kim	290	23	level	level	NN
kim	290	24	of	of	IN
kim	290	25	performance	performance	NN
kim	290	26	.	.	.
kim	291	1	This	this	DT
kim	291	2	is	be	VBZ
kim	291	3	particularly	particularly	RB
kim	291	4	so	so	RB
kim	291	5	if	if	IN
kim	291	6	the	the	DT
kim	291	7	pursuit	pursuit	NN
kim	291	8	of	of	IN
kim	291	9	such	such	JJ
kim	291	10	human-	human-	NN
kim	291	11	or	or	CC
kim	291	12	superhuman	superhuman	JJ
kim	291	13	-	-	HYPH
kim	291	14	level	level	NN
kim	291	15	performance	performance	NN
kim	291	16	is	be	VBZ
kim	291	17	likely	likely	JJ
kim	291	18	to	to	TO
kim	291	19	increase	increase	VB
kim	291	20	the	the	DT
kim	291	21	possibility	possibility	NN
kim	291	22	of	of	IN
kim	291	23	the	the	DT
kim	291	24	AI	AI	NNP
kim	291	25	algorithms	algorithm	NNS
kim	291	26	making	make	VBG
kim	291	27	unethical	unethical	JJ
kim	291	28	decisions	decision	NNS
kim	291	29	that	that	WDT
kim	291	30	negatively	negatively	RB
kim	291	31	impact	impact	VBP
kim	291	32	a	a	DT
kim	291	33	significant	significant	JJ
kim	291	34	number	number	NN
kim	291	35	of	of	IN
kim	291	36	people	people	NNS
kim	291	37	.	.	.
kim	292	1	We	-PRON-	PRP
kim	292	2	do	do	VBP
kim	292	3	not	not	RB
kim	292	4	have	have	VB
kim	292	5	to	to	TO
kim	292	6	task	task	NN
kim	292	7	AI	AI	NNP
kim	292	8	with	with	IN
kim	292	9	automating	automate	VBG
kim	292	10	away	away	RP
kim	292	11	human	human	JJ
kim	292	12	work	work	NN
kim	292	13	and	and	CC
kim	292	14	decisions	decision	NNS
kim	292	15	as	as	RB
kim	292	16	much	much	RB
kim	292	17	as	as	IN
kim	292	18	possible	possible	JJ
kim	292	19	.	.	.
kim	293	1	What	what	WP
kim	293	2	if	if	IN
kim	293	3	we	-PRON-	PRP
kim	293	4	reframe	reframe	VBP
kim	293	5	AI	AI	NNP
kim	293	6	’s	’s	POS
kim	293	7	role	role	NN
kim	293	8	in	in	IN
kim	293	9	the	the	DT
kim	293	10	context	context	NN
kim	293	11	of	of	IN
kim	293	12	assisting	assist	VBG
kim	293	13	people	people	NNS
kim	293	14	rather	rather	RB
kim	293	15	than	than	IN
kim	293	16	replacing	replace	VBG
kim	293	17	them	-PRON-	PRP
kim	293	18	,	,	,
kim	293	19	so	so	IN
kim	293	20	that	that	IN
kim	293	21	people	people	NNS
kim	293	22	can	can	MD
kim	293	23	become	become	VB
kim	293	24	more	more	RBR
kim	293	25	intelligent	intelligent	JJ
kim	293	26	and	and	CC
kim	293	27	more	more	RBR
kim	293	28	capable	capable	JJ
kim	293	29	where	where	WRB
kim	293	30	they	-PRON-	PRP
kim	293	31	struggle	struggle	VBP
kim	293	32	or	or	CC
kim	293	33	experience	experience	NN
kim	293	34	disadvantages	disadvantage	NNS
kim	293	35	,	,	,
kim	293	36	such	such	JJ
kim	293	37	as	as	IN
kim	293	38	critical	critical	JJ
kim	293	39	thinking	thinking	NN
kim	293	40	,	,	,
kim	293	41	civic	civic	JJ
kim	293	42	participation	participation	NN
kim	293	43	,	,	,
kim	293	44	healthy	healthy	JJ
kim	293	45	living	living	NN
kim	293	46	,	,	,
kim	293	47	financial	financial	JJ
kim	293	48	literacy	literacy	NN
kim	293	49	,	,	,
kim	293	50	dyslexia	dyslexia	NNP
kim	293	51	,	,	,
kim	293	52	or	or	CC
kim	293	53	hearing	hearing	NN
kim	293	54	loss	loss	NN
kim	293	55	?	?	.
kim	294	1	What	what	WDT
kim	294	2	kind	kind	NN
kim	294	3	of	of	IN
kim	294	4	AI	ai	NN
kim	294	5	-	-	HYPH
kim	294	6	driven	drive	VBN
kim	294	7	information	information	NN
kim	294	8	systems	system	NNS
kim	294	9	and	and	CC
kim	294	10	environments	environment	NNS
kim	294	11	would	would	MD
kim	294	12	be	be	VB
kim	294	13	created	create	VBN
kim	294	14	if	if	IN
kim	294	15	libraries	library	NNS
kim	294	16	approach	approach	VBP
kim	294	17	AI	AI	NNP
kim	294	18	with	with	IN
kim	294	19	such	such	JJ
kim	294	20	intention	intention	NN
kim	294	21	from	from	IN
kim	294	22	the	the	DT
kim	294	23	beginning	beginning	NN
kim	294	24	?	?	.
kim	295	1	References	References	NNP
kim	295	2	Alexander	Alexander	NNP
kim	295	3	,	,	,
kim	295	4	Larry	Larry	NNP
kim	295	5	,	,	,
kim	295	6	and	and	CC
kim	295	7	Michael	Michael	NNP
kim	295	8	Moore	Moore	NNP
kim	295	9	.	.	.
kim	296	1	2016	2016	CD
kim	296	2	.	.	.
kim	297	1	“	"	``
kim	297	2	Deontological	Deontological	NNP
kim	297	3	Ethics	Ethics	NNPS
kim	297	4	.	.	.
kim	297	5	”	"	''
kim	297	6	In	in	IN
kim	297	7	The	the	DT
kim	297	8	Stanford	Stanford	NNP
kim	297	9	Encyclopedia	Encyclopedia	NNP
kim	297	10	of	of	IN
kim	297	11	Philosophy	Philosophy	NNP
kim	297	12	,	,	,
kim	297	13	edited	edit	VBN
kim	297	14	by	by	IN
kim	297	15	Edward	Edward	NNP
kim	297	16	N.	N.	NNP
kim	297	17	Zalta	Zalta	NNP
kim	297	18	,	,	,
kim	297	19	Winter	Winter	NNP
kim	297	20	2016	2016	CD
kim	297	21	.	.	.
kim	298	1	Metaphysics	Metaphysics	NNP
kim	298	2	Research	Research	NNP
kim	298	3	Lab	Lab	NNP
kim	298	4	,	,	,
kim	298	5	Stanford	Stanford	NNP
kim	298	6	University	University	NNP
kim	298	7	.	.	.
kim	299	1	https://plato.stanford.edu/archives/win2016/entries/ethics-deontological/.	https://plato.stanford.edu/archives/win2016/entries/ethics-deontological/.	NNP
kim	300	1	Amini	Amini	NNP
kim	300	2	,	,	,
kim	300	3	Alexander	Alexander	NNP
kim	300	4	,	,	,
kim	300	5	Ava	Ava	NNP
kim	300	6	P.	P.	NNP
kim	300	7	Soleimany	Soleimany	NNP
kim	300	8	,	,	,
kim	300	9	Wilko	Wilko	NNP
kim	300	10	Schwarting	Schwarting	NNP
kim	300	11	,	,	,
kim	300	12	Sangeeta	Sangeeta	NNP
kim	300	13	N.	N.	NNP
kim	300	14	Bhatia	Bhatia	NNP
kim	300	15	,	,	,
kim	300	16	and	and	CC
kim	300	17	Daniela	Daniela	NNP
kim	300	18	Rus	Rus	NNP
kim	300	19	.	.	.
kim	301	1	2019	2019	CD
kim	301	2	.	.	.
kim	302	1	“	"	``
kim	302	2	Uncovering	uncover	VBG
kim	302	3	and	and	CC
kim	302	4	Mitigating	Mitigating	NNP
kim	302	5	Algorithmic	Algorithmic	NNP
kim	302	6	Bias	Bias	NNP
kim	302	7	through	through	IN
kim	302	8	Learned	learn	VBN
kim	302	9	Latent	Latent	NNP
kim	302	10	Structure	Structure	NNP
kim	302	11	.	.	.
kim	302	12	”	"	''
kim	302	13	In	in	IN
kim	302	14	Proceedings	Proceedings	NNP
kim	302	15	of	of	IN
kim	302	16	the	the	DT
kim	302	17	2019	2019	CD
kim	302	18	AAAI	AAAI	NNP
kim	302	19	/	/	SYM
kim	302	20	ACM	ACM	NNP
kim	302	21	Conference	Conference	NNP
kim	302	22	on	on	IN
kim	302	23	AI	AI	NNP
kim	302	24	,	,	,
kim	302	25	Ethics	Ethics	NNP
kim	302	26	,	,	,
kim	302	27	and	and	CC
kim	302	28	Society	Society	NNP
kim	302	29	,	,	,
kim	302	30	289–295	289–295	CD
kim	302	31	.	.	.
kim	303	1	AIES	AIES	NNP
kim	303	2	’	’	POS
kim	303	3	19	19	CD
kim	303	4	.	.	.
kim	304	1	New	New	NNP
kim	304	2	York	York	NNP
kim	304	3	,	,	,
kim	304	4	NY	NY	NNP
kim	304	5	,	,	,
kim	304	6	USA	USA	NNP
kim	304	7	:	:	:
kim	304	8	Association	Association	NNP
kim	304	9	for	for	IN
kim	304	10	Computing	Computing	NNP
kim	304	11	Machinery	Machinery	NNP
kim	304	12	.	.	.
kim	305	1	https://doi.org/10.1145/3306618.3314243	https://doi.org/10.1145/3306618.3314243	ADD
kim	305	2	.	.	.
kim	306	1	Averkamp	Averkamp	NNP
kim	306	2	,	,	,
kim	306	3	Shawn	Shawn	NNP
kim	306	4	,	,	,
kim	306	5	and	and	CC
kim	306	6	Julie	Julie	NNP
kim	306	7	Hardesty	Hardesty	NNP
kim	306	8	.	.	.
kim	307	1	2020	2020	CD
kim	307	2	.	.	.
kim	308	1	“	"	``
kim	308	2	AI	AI	NNP
kim	308	3	Is	be	VBZ
kim	308	4	Such	such	PDT
kim	308	5	a	a	DT
kim	308	6	Tool	tool	NN
kim	308	7	:	:	:
kim	308	8	Keeping	keep	VBG
kim	308	9	Your	-PRON-	PRP$
kim	308	10	Machine	machine	NN
kim	308	11	Learning	Learning	NNP
kim	308	12	Outputs	Outputs	NNPS
kim	308	13	in	in	IN
kim	308	14	Check	Check	NNP
kim	308	15	.	.	.
kim	308	16	”	"	''
kim	308	17	Presented	present	VBN
kim	308	18	at	at	IN
kim	308	19	the	the	DT
kim	308	20	Code4lib	Code4lib	NNP
kim	308	21	Conference	Conference	NNP
kim	308	22	,	,	,
kim	308	23	Pittsburgh	Pittsburgh	NNP
kim	308	24	,	,	,
kim	308	25	PA	PA	NNP
kim	308	26	,	,	,
kim	308	27	March	March	NNP
kim	308	28	11	11	CD
kim	308	29	.	.	.
kim	308	30	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
kim	308	31	.	.	.
kim	309	1	Blewer	Blewer	NNP
kim	309	2	,	,	,
kim	309	3	Ashley	Ashley	NNP
kim	309	4	,	,	,
kim	309	5	Bohyun	Bohyun	NNP
kim	309	6	Kim	Kim	NNP
kim	309	7	,	,	,
kim	309	8	and	and	CC
kim	309	9	Eric	Eric	NNP
kim	309	10	Phetteplace	Phetteplace	NNP
kim	309	11	.	.	.
kim	310	1	2018	2018	CD
kim	310	2	.	.	.
kim	311	1	“	"	``
kim	311	2	Reflections	reflection	NNS
kim	311	3	on	on	IN
kim	311	4	Code4Lib	Code4Lib	NNP
kim	311	5	2018	2018	CD
kim	311	6	.	.	.
kim	311	7	”	"	''
kim	311	8	ACRL	ACRL	NNP
kim	311	9	TechConnect	TechConnect	NNP
kim	311	10	(	(	-LRB-
kim	311	11	blog	blog	NN
kim	311	12	)	)	-RRB-
kim	311	13	.	.	.
kim	312	1	March	March	NNP
kim	312	2	12	12	CD
kim	312	3	,	,	,
kim	312	4	2018	2018	CD
kim	312	5	.	.	.
kim	312	6	https://acrl.ala.org/techconnect/post/reflections-on-code4lib-2018/.	https://acrl.ala.org/techconnect/post/reflections-on-code4lib-2018/.	NNP
kim	313	1	Boden	Boden	NNP
kim	313	2	,	,	,
kim	313	3	Margaret	Margaret	NNP
kim	313	4	A.	A.	NNP
kim	314	1	2016	2016	CD
kim	314	2	.	.	.
kim	315	1	AI	ai	NN
kim	315	2	:	:	:
kim	315	3	Its	-PRON-	PRP$
kim	315	4	Nature	Nature	NNP
kim	315	5	and	and	CC
kim	315	6	Future	Future	NNP
kim	315	7	.	.	.
kim	316	1	Oxford	oxford	NN
kim	316	2	:	:	:
kim	316	3	Oxford	Oxford	NNP
kim	316	4	University	University	NNP
kim	316	5	Press	Press	NNP
kim	316	6	.	.	.
kim	317	1	Bonnefon	Bonnefon	NNP
kim	317	2	,	,	,
kim	317	3	Jean	Jean	NNP
kim	317	4	-	-	HYPH
kim	317	5	François	François	NNP
kim	317	6	,	,	,
kim	317	7	Azim	Azim	NNP
kim	317	8	Shariff	Shariff	NNP
kim	317	9	,	,	,
kim	317	10	and	and	CC
kim	317	11	Iyad	Iyad	NNP
kim	317	12	Rahwan	Rahwan	NNP
kim	317	13	.	.	.
kim	318	1	2016	2016	CD
kim	318	2	.	.	.
kim	319	1	“	"	``
kim	319	2	The	the	DT
kim	319	3	Social	Social	NNP
kim	319	4	Dilemma	Dilemma	NNP
kim	319	5	of	of	IN
kim	319	6	Autonomous	Autonomous	NNP
kim	319	7	Vehicles	Vehicles	NNPS
kim	319	8	.	.	.
kim	319	9	”	"	''
kim	319	10	Science	science	NN
kim	319	11	352	352	CD
kim	319	12	(	(	-LRB-
kim	319	13	6293	6293	CD
kim	319	14	)	)	-RRB-
kim	319	15	:	:	:
kim	319	16	1573–76	1573–76	LS
kim	319	17	.	.	.
kim	320	1	https://doi.org/10.1126/science.aaf2654	https://doi.org/10.1126/science.aaf2654	NNP
kim	320	2	.	.	.
kim	321	1	Clark	Clark	NNP
kim	321	2	,	,	,
kim	321	3	Liat	Liat	NNP
kim	321	4	.	.	.
kim	322	1	2012	2012	CD
kim	322	2	.	.	.
kim	323	1	“	"	``
kim	323	2	Google	Google	NNP
kim	323	3	’s	’s	NNP
kim	323	4	Artificial	Artificial	NNP
kim	323	5	Brain	Brain	NNP
kim	323	6	Learns	learn	VBZ
kim	323	7	to	to	TO
kim	323	8	Find	find	VB
kim	323	9	Cat	cat	NN
kim	323	10	Videos	video	NNS
kim	323	11	.	.	.
kim	323	12	”	"	''
kim	323	13	Wired	wire	VBN
kim	323	14	,	,	,
kim	323	15	June	June	NNP
kim	323	16	26	26	CD
kim	323	17	,	,	,
kim	323	18	2012	2012	CD
kim	323	19	.	.	.
kim	323	20	https://www.wired.com/2012/06/google-x-neural-network/.	https://www.wired.com/2012/06/google-x-neural-network/.	NNP
kim	324	1	Conitzer	Conitzer	NNP
kim	324	2	,	,	,
kim	324	3	Vincent	Vincent	NNP
kim	324	4	,	,	,
kim	324	5	Walter	Walter	NNP
kim	324	6	Sinnott	Sinnott	NNP
kim	324	7	-	-	HYPH
kim	324	8	Armstrong	Armstrong	NNP
kim	324	9	,	,	,
kim	324	10	Jana	Jana	NNP
kim	324	11	Schaich	Schaich	NNP
kim	324	12	Borg	Borg	NNP
kim	324	13	,	,	,
kim	324	14	Yuan	Yuan	NNP
kim	324	15	Deng	Deng	NNP
kim	324	16	,	,	,
kim	324	17	and	and	CC
kim	324	18	Max	Max	NNP
kim	324	19	Kramer	Kramer	NNP
kim	324	20	.	.	.
kim	325	1	2017	2017	CD
kim	325	2	.	.	.
kim	326	1	“	"	``
kim	326	2	Moral	moral	JJ
kim	326	3	Decision	decision	NN
kim	326	4	Making	make	VBG
kim	326	5	Frameworks	Frameworks	NNP
kim	326	6	for	for	IN
kim	326	7	Artificial	Artificial	NNP
kim	326	8	Intelligence	Intelligence	NNP
kim	326	9	.	.	.
kim	326	10	”	"	''
kim	326	11	In	in	IN
kim	326	12	Proceedings	Proceedings	NNP
kim	326	13	of	of	IN
kim	326	14	the	the	DT
kim	326	15	Thirty	Thirty	NNP
kim	326	16	-	-	HYPH
kim	326	17	First	First	NNP
kim	326	18	AAAI	AAAI	NNP
kim	326	19	Conference	Conference	NNP
kim	326	20	on	on	IN
kim	326	21	Artificial	Artificial	NNP
kim	326	22	Intelligence	Intelligence	NNP
kim	326	23	,	,	,
kim	326	24	4831–4835	4831–4835	CD
kim	326	25	.	.	.
kim	327	1	AAAI’17	aaai’17	UH
kim	327	2	.	.	.
kim	328	1	San	San	NNP
kim	328	2	Francisco	Francisco	NNP
kim	328	3	,	,	,
kim	328	4	California	California	NNP
kim	328	5	,	,	,
kim	328	6	USA	USA	NNP
kim	328	7	:	:	:
kim	328	8	AAAI	AAAI	NNP
kim	328	9	Press	Press	NNP
kim	328	10	.	.	.
kim	329	1	Courtland	Courtland	NNP
kim	329	2	,	,	,
kim	329	3	Rachel	Rachel	NNP
kim	329	4	.	.	.
kim	330	1	2018	2018	CD
kim	330	2	.	.	.
kim	331	1	“	"	``
kim	331	2	Bias	Bias	NNP
kim	331	3	Detectives	detective	NNS
kim	331	4	:	:	:
kim	331	5	The	the	DT
kim	331	6	Researchers	researcher	NNS
kim	331	7	Striving	strive	VBG
kim	331	8	to	to	TO
kim	331	9	Make	make	VB
kim	331	10	Algorithms	Algorithms	NNP
kim	331	11	Fair	Fair	NNP
kim	331	12	.	.	.
kim	331	13	”	"	''
kim	331	14	Nature	nature	NN
kim	331	15	558	558	CD
kim	331	16	(	(	-LRB-
kim	331	17	7710	7710	CD
kim	331	18	)	)	-RRB-
kim	331	19	:	:	:
kim	331	20	357–60	357–60	CD
kim	331	21	.	.	.
kim	332	1	https://doi.org/10.1038/d41586-018-05469-3	https://doi.org/10.1038/d41586-018-05469-3	ADD
kim	332	2	.	.	.
kim	333	1	Cushman	Cushman	NNP
kim	333	2	,	,	,
kim	333	3	Fiery	Fiery	NNP
kim	333	4	,	,	,
kim	333	5	Liane	Liane	NNP
kim	333	6	Young	Young	NNP
kim	333	7	,	,	,
kim	333	8	and	and	CC
kim	333	9	Marc	Marc	NNP
kim	333	10	Hauser	Hauser	NNP
kim	333	11	.	.	.
kim	334	1	2006	2006	CD
kim	334	2	.	.	.
kim	335	1	“	"	``
kim	335	2	The	the	DT
kim	335	3	Role	Role	NNP
kim	335	4	of	of	IN
kim	335	5	Conscious	Conscious	NNP
kim	335	6	Reasoning	Reasoning	NNP
kim	335	7	and	and	CC
kim	335	8	Intuition	Intuition	NNP
kim	335	9	in	in	IN
kim	335	10	Moral	Moral	NNP
kim	335	11	Judgment	Judgment	NNP
kim	335	12	:	:	:
kim	335	13	Testing	test	VBG
kim	335	14	Three	three	CD
kim	335	15	Principles	principle	NNS
kim	335	16	of	of	IN
kim	335	17	Harm	harm	NN
kim	335	18	.	.	.
kim	335	19	”	"	''
kim	335	20	Psychological	Psychological	NNP
kim	335	21	Science	Science	NNP
kim	335	22	17	17	CD
kim	335	23	(	(	-LRB-
kim	335	24	12	12	CD
kim	335	25	)	)	-RRB-
kim	335	26	:	:	:
kim	335	27	1082–89	1082–89	LS
kim	335	28	.	.	.
kim	336	1	Davis	Davis	NNP
kim	336	2	,	,	,
kim	336	3	Daniel	Daniel	NNP
kim	336	4	L.	L.	NNP
kim	336	5	2007	2007	CD
kim	336	6	.	.	.
kim	337	1	“	"	``
kim	337	2	Who	who	WP
kim	337	3	Decides	decide	VBZ
kim	337	4	:	:	:
kim	337	5	Man	man	NN
kim	337	6	or	or	CC
kim	337	7	Machine	machine	NN
kim	337	8	?	?	.
kim	337	9	”	"	''
kim	337	10	Armed	Armed	NNP
kim	337	11	Forces	Forces	NNPS
kim	337	12	Journal	Journal	NNP
kim	337	13	,	,	,
kim	337	14	November	November	NNP
kim	337	15	.	.	.
kim	338	1	http://armedforcesjournal.com/who-decides-man-or-machine/.	http://armedforcesjournal.com/who-decides-man-or-machine/.	NNP
kim	339	1	Davis	Davis	NNP
kim	339	2	,	,	,
kim	339	3	Hannah	Hannah	NNP
kim	339	4	.	.	.
kim	340	1	2020	2020	CD
kim	340	2	.	.	.
kim	341	1	“	"	``
kim	341	2	A	a	DT
kim	341	3	Dataset	dataset	NN
kim	341	4	Is	be	VBZ
kim	341	5	a	a	DT
kim	341	6	Worldview	worldview	NN
kim	341	7	.	.	.
kim	341	8	”	"	''
kim	341	9	Towards	towards	IN
kim	341	10	Data	Data	NNP
kim	341	11	Science	Science	NNP
kim	341	12	.	.	.
kim	342	1	March	March	NNP
kim	342	2	5	5	CD
kim	342	3	,	,	,
kim	342	4	2020	2020	CD
kim	342	5	.	.	.
kim	342	6	https://towardsdatascience.com/a-dataset-is-a-worldview-5328216dd44d	https://towardsdatascience.com/a-dataset-is-a-worldview-5328216dd44d	ADD
kim	342	7	.	.	.
kim	343	1	Foot	Foot	NNP
kim	343	2	,	,	,
kim	343	3	Philippa	Philippa	NNP
kim	343	4	.	.	.
kim	344	1	1967	1967	CD
kim	344	2	.	.	.
kim	345	1	“	"	``
kim	345	2	The	the	DT
kim	345	3	Problem	problem	NN
kim	345	4	of	of	IN
kim	345	5	Abortion	abortion	NN
kim	345	6	and	and	CC
kim	345	7	the	the	DT
kim	345	8	Doctrine	Doctrine	NNP
kim	345	9	of	of	IN
kim	345	10	Double	Double	NNP
kim	345	11	Effect	Effect	NNP
kim	345	12	.	.	.
kim	345	13	”	"	''
kim	345	14	Oxford	Oxford	NNP
kim	345	15	Review	Review	NNP
kim	345	16	5	5	CD
kim	345	17	:	:	:
kim	345	18	5–15	5–15	CD
kim	345	19	.	.	.
kim	346	1	Heaven	Heaven	NNP
kim	346	2	,	,	,
kim	346	3	Will	Will	NNP
kim	346	4	Douglas	Douglas	NNP
kim	346	5	.	.	.
kim	347	1	2020	2020	CD
kim	347	2	.	.	.
kim	348	1	“	"	``
kim	348	2	DeepMind	DeepMind	NNP
kim	348	3	’s	’s	,
kim	348	4	AI	ai	NN
kim	348	5	Can	Can	MD
kim	348	6	Now	now	RB
kim	348	7	Play	play	VB
kim	348	8	All	all	DT
kim	348	9	57	57	CD
kim	348	10	Atari	Atari	NNP
kim	348	11	Games	Games	NNPS
kim	348	12	—	—	:
kim	348	13	but	but	CC
kim	348	14	It	-PRON-	PRP
kim	348	15	’s	’	VBZ
kim	348	16	Still	still	RB
kim	348	17	Not	not	RB
kim	348	18	Versatile	versatile	JJ
kim	348	19	Enough	enough	RB
kim	348	20	.	.	.
kim	348	21	”	"	''
kim	348	22	MIT	MIT	NNP
kim	348	23	Technology	Technology	NNP
kim	348	24	Review	Review	NNP
kim	348	25	,	,	,
kim	348	26	April	April	NNP
kim	348	27	1	1	CD
kim	348	28	,	,	,
kim	348	29	2020	2020	CD
kim	348	30	.	.	.
kim	348	31	https://www.technologyreview.com/2020/04/01/974997/deepminds-ai-57-atari-games-but-its-still-not-versatile-enough/.	https://www.technologyreview.com/2020/04/01/974997/deepminds-ai-57-atari-games-but-its-still-not-versatile-enough/.	NNP
kim	349	1	International	International	NNP
kim	349	2	Committee	Committee	NNP
kim	349	3	of	of	IN
kim	349	4	the	the	DT
kim	349	5	Red	Red	NNP
kim	349	6	Cross	Cross	NNP
kim	349	7	.	.	.
kim	350	1	2015	2015	CD
kim	350	2	.	.	.
kim	351	1	“	"	``
kim	351	2	What	what	WP
kim	351	3	Are	be	VBP
kim	351	4	Jus	Jus	NNP
kim	351	5	Ad	Ad	NNP
kim	351	6	Bellum	Bellum	NNP
kim	351	7	and	and	CC
kim	351	8	Jus	Jus	NNP
kim	351	9	in	in	IN
kim	351	10	Bello	Bello	NNP
kim	351	11	?	?	.
kim	351	12	,	,	,
kim	351	13	”	"	''
kim	351	14	September	September	NNP
kim	351	15	.	.	.
kim	352	1	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
kim	352	2	.	.	.
kim	353	1	Kahn	Kahn	NNP
kim	353	2	,	,	,
kim	353	3	Leonard	Leonard	NNP
kim	353	4	.	.	.
kim	354	1	2012	2012	CD
kim	354	2	.	.	.
kim	355	1	“	"	``
kim	355	2	Military	military	JJ
kim	355	3	Robots	Robots	NNP
kim	355	4	and	and	CC
kim	355	5	The	the	DT
kim	355	6	Likelihood	Likelihood	NNP
kim	355	7	of	of	IN
kim	355	8	Armed	Armed	NNP
kim	355	9	Combat	Combat	NNP
kim	355	10	.	.	.
kim	355	11	”	"	''
kim	355	12	In	in	IN
kim	355	13	Robot	Robot	NNP
kim	355	14	Ethics	Ethics	NNP
kim	355	15	:	:	:
kim	355	16	The	the	DT
kim	355	17	Ethical	Ethical	NNP
kim	355	18	and	and	CC
kim	355	19	Social	Social	NNP
kim	355	20	Implications	Implications	NNPS
kim	355	21	of	of	IN
kim	355	22	Robotics	Robotics	NNPS
kim	355	23	,	,	,
kim	355	24	edited	edit	VBN
kim	355	25	by	by	IN
kim	355	26	Patrick	Patrick	NNP
kim	355	27	Lin	Lin	NNP
kim	355	28	,	,	,
kim	355	29	Keith	Keith	NNP
kim	355	30	Abney	Abney	NNP
kim	355	31	,	,	,
kim	355	32	and	and	CC
kim	355	33	George	George	NNP
kim	355	34	A.	A.	NNP
kim	355	35	Bekey	Bekey	NNP
kim	355	36	,	,	,
kim	355	37	274–92	274–92	CD
kim	355	38	.	.	.
kim	356	1	Intelligent	intelligent	JJ
kim	356	2	Robotics	Robotics	NNPS
kim	356	3	and	and	CC
kim	356	4	Autonomous	Autonomous	NNP
kim	356	5	Agents	Agents	NNPS
kim	356	6	.	.	.
kim	357	1	Cambridge	Cambridge	NNP
kim	357	2	,	,	,
kim	357	3	Mass.	Massachusetts	NNP
kim	358	1	:	:	:
kim	358	2	MIT	MIT	NNP
kim	358	3	Press	Press	NNP
kim	358	4	.	.	.
kim	359	1	Knight	Knight	NNP
kim	359	2	,	,	,
kim	359	3	Will	Will	NNP
kim	359	4	.	.	.
kim	360	1	2017	2017	CD
kim	360	2	.	.	.
kim	361	1	“	"	``
kim	361	2	The	the	DT
kim	361	3	Dark	Dark	NNP
kim	361	4	Secret	Secret	NNP
kim	361	5	at	at	IN
kim	361	6	the	the	DT
kim	361	7	Heart	heart	NN
kim	361	8	of	of	IN
kim	361	9	AI	AI	NNP
kim	361	10	.	.	.
kim	361	11	”	"	''
kim	361	12	MIT	MIT	NNP
kim	361	13	Technology	Technology	NNP
kim	361	14	Review	Review	NNP
kim	361	15	,	,	,
kim	361	16	April	April	NNP
kim	361	17	11	11	CD
kim	361	18	,	,	,
kim	361	19	2017	2017	CD
kim	361	20	.	.	.
kim	361	21	https://www.technologyreview.com/2017/04/11/5113/the-dark-secret-at-the-heart-of-ai/.	https://www.technologyreview.com/2017/04/11/5113/the-dark-secret-at-the-heart-of-ai/.	CD
kim	362	1	—	—	:
kim	362	2	—	—	:
kim	362	3	—	—	:
kim	362	4	.	.	.
kim	363	1	2019a	2019a	CD
kim	363	2	.	.	.
kim	364	1	“	"	``
kim	364	2	Two	two	CD
kim	364	3	Rival	rival	JJ
kim	364	4	AI	AI	NNP
kim	364	5	Approaches	approach	NNS
kim	364	6	Combine	combine	VBP
kim	364	7	to	to	TO
kim	364	8	Let	let	VB
kim	364	9	Machines	machine	NNS
kim	364	10	Learn	learn	VB
kim	364	11	about	about	IN
kim	364	12	the	the	DT
kim	364	13	World	World	NNP
kim	364	14	like	like	IN
kim	364	15	a	a	DT
kim	364	16	Child	child	NN
kim	364	17	.	.	.
kim	364	18	”	"	''
kim	364	19	MIT	MIT	NNP
kim	364	20	Technology	Technology	NNP
kim	364	21	Review	Review	NNP
kim	364	22	,	,	,
kim	364	23	April	April	NNP
kim	364	24	8	8	CD
kim	364	25	,	,	,
kim	364	26	2019	2019	CD
kim	364	27	.	.	.
kim	364	28	https://www.technologyreview.com/2019/04/08/103223/two-rival-ai-approaches-combine-to-let-machines-learn-about-the-world-like-a-child/.	https://www.technologyreview.com/2019/04/08/103223/two-rival-ai-approaches-combine-to-let-machines-learn-about-the-world-like-a-child/.	NNP
kim	365	1	—	—	:
kim	365	2	—	—	:
kim	365	3	—	—	:
kim	365	4	.	.	.
kim	366	1	2019b	2019b	CD
kim	366	2	.	.	.
kim	367	1	“	"	``
kim	367	2	AI	AI	NNP
kim	367	3	Is	be	VBZ
kim	367	4	Biased	bias	VBN
kim	367	5	.	.	.
kim	368	1	Here	here	RB
kim	368	2	’s	’	VBZ
kim	368	3	How	how	WRB
kim	368	4	Scientists	scientist	NNS
kim	368	5	Are	be	VBP
kim	368	6	Trying	try	VBG
kim	368	7	to	to	TO
kim	368	8	Fix	fix	VB
kim	368	9	It	-PRON-	PRP
kim	368	10	.	.	.
kim	368	11	”	"	''
kim	368	12	Wired	wire	VBN
kim	368	13	,	,	,
kim	368	14	December	December	NNP
kim	368	15	19	19	CD
kim	368	16	,	,	,
kim	368	17	2019	2019	CD
kim	368	18	.	.	.
kim	368	19	https://www.wired.com/story/ai-biased-how-scientists-trying-fix/.	https://www.wired.com/story/ai-biased-how-scientists-trying-fix/.	NNP
kim	369	1	Koch	Koch	NNP
kim	369	2	,	,	,
kim	369	3	Christof	Christof	NNP
kim	369	4	.	.	.
kim	370	1	2016	2016	CD
kim	370	2	.	.	.
kim	371	1	“	"	``
kim	371	2	How	how	WRB
kim	371	3	the	the	DT
kim	371	4	Computer	Computer	NNP
kim	371	5	Beat	beat	VBP
kim	371	6	the	the	DT
kim	371	7	Go	Go	NNP
kim	371	8	Master	Master	NNP
kim	371	9	.	.	.
kim	371	10	”	"	''
kim	371	11	Scientific	scientific	JJ
kim	371	12	American	American	NNP
kim	371	13	.	.	.
kim	372	1	March	March	NNP
kim	372	2	19	19	CD
kim	372	3	,	,	,
kim	372	4	2016	2016	CD
kim	372	5	.	.	.
kim	372	6	https://www.scientificamerican.com/article/how-the-computer-beat-the-go-master/.	https://www.scientificamerican.com/article/how-the-computer-beat-the-go-master/.	ADD
kim	373	1	Le	Le	NNP
kim	373	2	,	,	,
kim	373	3	Quoc	Quoc	NNP
kim	373	4	V.	V.	NNP
kim	373	5	,	,	,
kim	373	6	and	and	CC
kim	373	7	Mike	Mike	NNP
kim	373	8	Schuster	Schuster	NNP
kim	373	9	.	.	.
kim	374	1	2016	2016	CD
kim	374	2	.	.	.
kim	375	1	“	"	``
kim	375	2	A	a	DT
kim	375	3	Neural	Neural	NNP
kim	375	4	Network	Network	NNP
kim	375	5	for	for	IN
kim	375	6	Machine	Machine	NNP
kim	375	7	Translation	Translation	NNP
kim	375	8	,	,	,
kim	375	9	at	at	IN
kim	375	10	Production	Production	NNP
kim	375	11	Scale	Scale	NNP
kim	375	12	.	.	.
kim	375	13	”	"	''
kim	375	14	Google	Google	NNP
kim	375	15	AI	AI	NNP
kim	375	16	Blog	Blog	NNP
kim	375	17	(	(	-LRB-
kim	375	18	blog	blog	NN
kim	375	19	)	)	-RRB-
kim	375	20	.	.	.
kim	376	1	September	September	NNP
kim	376	2	27	27	CD
kim	376	3	,	,	,
kim	376	4	2016	2016	CD
kim	376	5	.	.	.
kim	376	6	http://ai.googleblog.com/2016/09/a-neural-network-for-machine.html	http://ai.googleblog.com/2016/09/a-neural-network-for-machine.html	NNP
kim	376	7	.	.	.
kim	377	1	Lohr	Lohr	NNP
kim	377	2	,	,	,
kim	377	3	Steve	Steve	NNP
kim	377	4	.	.	.
kim	378	1	2018	2018	CD
kim	378	2	.	.	.
kim	379	1	“	"	``
kim	379	2	Facial	facial	JJ
kim	379	3	Recognition	Recognition	NNP
kim	379	4	Is	be	VBZ
kim	379	5	Accurate	accurate	JJ
kim	379	6	,	,	,
kim	379	7	If	if	IN
kim	379	8	You	-PRON-	PRP
kim	379	9	’re	be	VBP
kim	379	10	a	a	DT
kim	379	11	White	White	NNP
kim	379	12	Guy	Guy	NNP
kim	379	13	.	.	.
kim	379	14	”	"	''
kim	379	15	New	New	NNP
kim	379	16	York	York	NNP
kim	379	17	Times	Times	NNP
kim	379	18	,	,	,
kim	379	19	February	February	NNP
kim	379	20	9	9	CD
kim	379	21	,	,	,
kim	379	22	2018	2018	CD
kim	379	23	.	.	.
kim	379	24	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	NNP
kim	379	25	.	.	.
kim	380	1	Mahlangu	Mahlangu	NNP
kim	380	2	,	,	,
kim	380	3	Isaac	Isaac	NNP
kim	380	4	.	.	.
kim	381	1	2019	2019	CD
kim	381	2	.	.	.
kim	382	1	“	"	``
kim	382	2	Meet	Meet	NNP
kim	382	3	Libby	Libby	NNP
kim	382	4	-	-	,
kim	382	5	the	the	DT
kim	382	6	New	New	NNP
kim	382	7	Robot	Robot	NNP
kim	382	8	Library	Library	NNP
kim	382	9	Assistant	Assistant	NNP
kim	382	10	at	at	IN
kim	382	11	the	the	DT
kim	382	12	University	University	NNP
kim	382	13	of	of	IN
kim	382	14	Pretoria	Pretoria	NNP
kim	382	15	’s	’s	NNP
kim	382	16	Hatfield	Hatfield	NNP
kim	382	17	Campus	Campus	NNP
kim	382	18	.	.	.
kim	382	19	”	"	''
kim	382	20	SowetanLIVE	sowetanlive	CD
kim	382	21	.	.	.
kim	383	1	June	June	NNP
kim	383	2	4	4	CD
kim	383	3	,	,	,
kim	383	4	2019	2019	CD
kim	383	5	.	.	.
kim	383	6	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/.	.
kim	384	1	Markoff	Markoff	NNP
kim	384	2	,	,	,
kim	384	3	John	John	NNP
kim	384	4	.	.	.
kim	385	1	2012	2012	CD
kim	385	2	.	.	.
kim	386	1	“	"	``
kim	386	2	How	how	WRB
kim	386	3	Many	many	JJ
kim	386	4	Computers	computer	NNS
kim	386	5	to	to	TO
kim	386	6	Identify	identify	VB
kim	386	7	a	a	DT
kim	386	8	Cat	cat	NN
kim	386	9	?	?	.
kim	387	1	16,000	16,000	CD
kim	387	2	.	.	.
kim	387	3	”	"	''
kim	387	4	New	New	NNP
kim	387	5	York	York	NNP
kim	387	6	Times	Times	NNP
kim	387	7	,	,	,
kim	387	8	June	June	NNP
kim	387	9	25	25	CD
kim	387	10	,	,	,
kim	387	11	2012	2012	CD
kim	387	12	.	.	.
kim	388	1	Marr	Marr	NNP
kim	388	2	,	,	,
kim	388	3	Bernard	Bernard	NNP
kim	388	4	.	.	.
kim	389	1	2018	2018	CD
kim	389	2	.	.	.
kim	390	1	“	"	``
kim	390	2	How	how	WRB
kim	390	3	AI	AI	NNP
kim	390	4	And	and	CC
kim	390	5	Machine	Machine	NNP
kim	390	6	Learning	Learning	NNP
kim	390	7	Are	be	VBP
kim	390	8	Transforming	transform	VBG
kim	390	9	Law	Law	NNP
kim	390	10	Firms	firm	NNS
kim	390	11	And	and	CC
kim	390	12	The	the	DT
kim	390	13	Legal	Legal	NNP
kim	390	14	Sector	Sector	NNP
kim	390	15	.	.	.
kim	390	16	”	"	''
kim	390	17	Forbes	Forbes	NNP
kim	390	18	,	,	,
kim	390	19	May	May	NNP
kim	390	20	23	23	CD
kim	390	21	,	,	,
kim	390	22	2018	2018	CD
kim	390	23	.	.	.
kim	390	24	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
kim	391	1	Pariser	Pariser	NNP
kim	391	2	,	,	,
kim	391	3	Eli	Eli	NNP
kim	391	4	.	.	.
kim	392	1	2011	2011	CD
kim	392	2	.	.	.
kim	393	1	The	the	DT
kim	393	2	Filter	Filter	NNP
kim	393	3	Bubble	bubble	NN
kim	393	4	:	:	:
kim	393	5	How	how	WRB
kim	393	6	the	the	DT
kim	393	7	New	New	NNP
kim	393	8	Personalized	Personalized	NNP
kim	393	9	Web	web	NN
kim	393	10	Is	be	VBZ
kim	393	11	Changing	change	VBG
kim	393	12	What	what	WP
kim	393	13	We	-PRON-	PRP
kim	393	14	Read	read	VBP
kim	393	15	and	and	CC
kim	393	16	How	how	WRB
kim	393	17	We	-PRON-	PRP
kim	393	18	Think	think	VBP
kim	393	19	.	.	.
kim	394	1	New	New	NNP
kim	394	2	York	York	NNP
kim	394	3	:	:	:
kim	394	4	Penguin	Penguin	NNP
kim	394	5	Press	Press	NNP
kim	394	6	.	.	.
kim	395	1	Porter	Porter	NNP
kim	395	2	,	,	,
kim	395	3	Jon	Jon	NNP
kim	395	4	.	.	.
kim	396	1	2018	2018	CD
kim	396	2	.	.	.
kim	397	1	“	"	``
kim	397	2	The	the	DT
kim	397	3	Pixel	Pixel	NNP
kim	397	4	Buds	Buds	NNPS
kim	397	5	’	’	POS
kim	397	6	Translation	translation	NN
kim	397	7	Feature	feature	NN
kim	397	8	Is	be	VBZ
kim	397	9	Coming	come	VBG
kim	397	10	to	to	IN
kim	397	11	All	all	DT
kim	397	12	Headphones	Headphones	NNPS
kim	397	13	with	with	IN
kim	397	14	Google	Google	NNP
kim	397	15	Assistant	Assistant	NNP
kim	397	16	.	.	.
kim	397	17	”	"	''
kim	397	18	The	the	DT
kim	397	19	Verge	Verge	NNP
kim	397	20	.	.	.
kim	398	1	October	October	NNP
kim	398	2	15	15	CD
kim	398	3	,	,	,
kim	398	4	2018	2018	CD
kim	398	5	.	.	.
kim	398	6	https://www.theverge.com/circuitbreaker/2018/10/15/17978298/pixel-buds-google-translate-google-assistant-headphones	https://www.theverge.com/circuitbreaker/2018/10/15/17978298/pixel-buds-google-translate-google-assistant-headphones	NNP
kim	398	7	.	.	.
kim	399	1	Price	price	NN
kim	399	2	,	,	,
kim	399	3	Gary	Gary	NNP
kim	399	4	.	.	.
kim	400	1	2019	2019	CD
kim	400	2	.	.	.
kim	401	1	“	"	``
kim	401	2	The	the	DT
kim	401	3	Library	Library	NNP
kim	401	4	of	of	IN
kim	401	5	Congress	Congress	NNP
kim	401	6	Posts	Posts	NNPS
kim	401	7	Solicitation	Solicitation	NNP
kim	401	8	For	for	IN
kim	401	9	a	a	DT
kim	401	10	Machine	Machine	NNP
kim	401	11	Learning	Learning	NNP
kim	401	12	/	/	SYM
kim	401	13	Deep	Deep	NNP
kim	401	14	Learning	Learning	NNP
kim	401	15	Pilot	Pilot	NNP
kim	401	16	Program	Program	NNP
kim	401	17	to	to	TO
kim	401	18	‘	'	``
kim	401	19	Maximize	maximize	VB
kim	401	20	the	the	DT
kim	401	21	Use	Use	NNP
kim	401	22	of	of	IN
kim	401	23	Its	-PRON-	PRP$
kim	401	24	Digital	Digital	NNP
kim	401	25	Collection	Collection	NNP
kim	401	26	.	.	.
kim	401	27	’	'	''
kim	401	28	”	"	''
kim	401	29	LJ	LJ	NNP
kim	401	30	InfoDOCKET	InfoDOCKET	NNP
kim	401	31	.	.	.
kim	402	1	June	June	NNP
kim	402	2	13	13	CD
kim	402	3	,	,	,
kim	402	4	2019	2019	CD
kim	402	5	.	.	.
kim	402	6	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
kim	403	1	Sharkey	Sharkey	NNP
kim	403	2	,	,	,
kim	403	3	Noel	Noel	NNP
kim	403	4	.	.	.
kim	404	1	2012	2012	CD
kim	404	2	.	.	.
kim	405	1	“	"	``
kim	405	2	Killing	kill	VBG
kim	405	3	Made	make	VBN
kim	405	4	Easy	Easy	NNP
kim	405	5	:	:	:
kim	405	6	From	from	IN
kim	405	7	Joysticks	Joysticks	NNP
kim	405	8	to	to	IN
kim	405	9	Politics	politic	NNS
kim	405	10	.	.	.
kim	405	11	”	"	''
kim	405	12	In	in	IN
kim	405	13	Robot	Robot	NNP
kim	405	14	Ethics	Ethics	NNP
kim	405	15	:	:	:
kim	405	16	The	the	DT
kim	405	17	Ethical	Ethical	NNP
kim	405	18	and	and	CC
kim	405	19	Social	Social	NNP
kim	405	20	Implications	Implications	NNPS
kim	405	21	of	of	IN
kim	405	22	Robotics	Robotics	NNPS
kim	405	23	,	,	,
kim	405	24	edited	edit	VBN
kim	405	25	by	by	IN
kim	405	26	Patrick	Patrick	NNP
kim	405	27	Lin	Lin	NNP
kim	405	28	,	,	,
kim	405	29	Keith	Keith	NNP
kim	405	30	Abney	Abney	NNP
kim	405	31	,	,	,
kim	405	32	and	and	CC
kim	405	33	George	George	NNP
kim	405	34	A.	A.	NNP
kim	405	35	Bekey	Bekey	NNP
kim	405	36	,	,	,
kim	405	37	111–28	111–28	CD
kim	405	38	.	.	.
kim	406	1	Intelligent	intelligent	JJ
kim	406	2	Robotics	Robotics	NNPS
kim	406	3	and	and	CC
kim	406	4	Autonomous	Autonomous	NNP
kim	406	5	Agents	Agents	NNPS
kim	406	6	.	.	.
kim	407	1	Cambridge	Cambridge	NNP
kim	407	2	,	,	,
kim	407	3	Mass.	Massachusetts	NNP
kim	408	1	:	:	:
kim	408	2	MIT	MIT	NNP
kim	408	3	Press	Press	NNP
kim	408	4	..	..	.
kim	408	5	Singer	singer	NN
kim	408	6	,	,	,
kim	408	7	Peter	Peter	NNP
kim	408	8	.	.	.
kim	409	1	2005	2005	CD
kim	409	2	.	.	.
kim	410	1	“	"	``
kim	410	2	Ethics	Ethics	NNPS
kim	410	3	and	and	CC
kim	410	4	Intuitions	Intuitions	NNPS
kim	410	5	.	.	.
kim	410	6	”	"	''
kim	410	7	The	the	DT
kim	410	8	Journal	Journal	NNP
kim	410	9	of	of	IN
kim	410	10	Ethics	Ethics	NNP
kim	410	11	9	9	CD
kim	410	12	(	(	-LRB-
kim	410	13	3/4	3/4	CD
kim	410	14	)	)	-RRB-
kim	410	15	:	:	:
kim	410	16	331–52	331–52	CD
kim	410	17	.	.	.
kim	411	1	Sinnott	Sinnott	NNP
kim	411	2	-	-	HYPH
kim	411	3	Armstrong	Armstrong	NNP
kim	411	4	,	,	,
kim	411	5	Walter	Walter	NNP
kim	411	6	.	.	.
kim	412	1	2019	2019	CD
kim	412	2	.	.	.
kim	413	1	“	"	``
kim	413	2	Consequentialism	consequentialism	NN
kim	413	3	.	.	.
kim	413	4	”	"	''
kim	413	5	In	in	IN
kim	413	6	The	the	DT
kim	413	7	Stanford	Stanford	NNP
kim	413	8	Encyclopedia	Encyclopedia	NNP
kim	413	9	of	of	IN
kim	413	10	Philosophy	Philosophy	NNP
kim	413	11	,	,	,
kim	413	12	edited	edit	VBN
kim	413	13	by	by	IN
kim	413	14	Edward	Edward	NNP
kim	413	15	N.	N.	NNP
kim	413	16	Zalta	Zalta	NNP
kim	413	17	,	,	,
kim	413	18	Summer	Summer	NNP
kim	413	19	2019	2019	CD
kim	413	20	.	.	.
kim	414	1	Metaphysics	Metaphysics	NNP
kim	414	2	Research	Research	NNP
kim	414	3	Lab	Lab	NNP
kim	414	4	,	,	,
kim	414	5	Stanford	Stanford	NNP
kim	414	6	University	University	NNP
kim	414	7	.	.	.
kim	415	1	https://plato.stanford.edu/archives/sum2019/entries/consequentialism/.	https://plato.stanford.edu/archives/sum2019/entries/consequentialism/.	ADD
kim	416	1	Stanley	Stanley	NNP
kim	416	2	,	,	,
kim	416	3	Jay	Jay	NNP
kim	416	4	.	.	.
kim	417	1	2017	2017	CD
kim	417	2	.	.	.
kim	418	1	“	"	``
kim	418	2	Pitfalls	pitfall	NNS
kim	418	3	of	of	IN
kim	418	4	Artificial	Artificial	NNP
kim	418	5	Intelligence	Intelligence	NNP
kim	418	6	Decisionmaking	Decisionmaking	NNP
kim	418	7	Highlighted	highlight	VBN
kim	418	8	In	in	IN
kim	418	9	Idaho	Idaho	NNP
kim	418	10	ACLU	ACLU	NNP
kim	418	11	Case	Case	NNP
kim	418	12	.	.	.
kim	418	13	”	"	''
kim	418	14	American	American	NNP
kim	418	15	Civil	Civil	NNP
kim	418	16	Liberties	Liberties	NNPS
kim	418	17	Union	Union	NNP
kim	418	18	(	(	-LRB-
kim	418	19	blog	blog	NN
kim	418	20	)	)	-RRB-
kim	418	21	.	.	.
kim	419	1	June	June	NNP
kim	419	2	2	2	CD
kim	419	3	,	,	,
kim	419	4	2017	2017	CD
kim	419	5	.	.	.
kim	419	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
kim	419	7	.	.	.
kim	420	1	Talley	Talley	NNP
kim	420	2	,	,	,
kim	420	3	Nancy	Nancy	NNP
kim	420	4	B.	B.	NNP
kim	421	1	2016	2016	CD
kim	421	2	.	.	.
kim	422	1	“	"	``
kim	422	2	Imagining	imagine	VBG
kim	422	3	the	the	DT
kim	422	4	Use	Use	NNP
kim	422	5	of	of	IN
kim	422	6	Intelligent	Intelligent	NNP
kim	422	7	Agents	Agents	NNPS
kim	422	8	and	and	CC
kim	422	9	Artificial	Artificial	NNP
kim	422	10	Intelligence	Intelligence	NNP
kim	422	11	in	in	IN
kim	422	12	Academic	Academic	NNP
kim	422	13	Law	Law	NNP
kim	422	14	Libraries	Libraries	NNPS
kim	422	15	.	.	.
kim	422	16	”	"	''
kim	422	17	Law	Law	NNP
kim	422	18	Library	Library	NNP
kim	422	19	Journal	Journal	NNP
kim	422	20	108	108	CD
kim	422	21	(	(	-LRB-
kim	422	22	3	3	CD
kim	422	23	)	)	-RRB-
kim	422	24	:	:	:
kim	422	25	383–402	383–402	CD
kim	422	26	.	.	.
kim	423	1	Tashea	Tashea	NNP
kim	423	2	,	,	,
kim	423	3	Jason	Jason	NNP
kim	423	4	.	.	.
kim	424	1	2017	2017	CD
kim	424	2	.	.	.
kim	425	1	“	"	``
kim	425	2	Courts	court	NNS
kim	425	3	Are	be	VBP
kim	425	4	Using	use	VBG
kim	425	5	AI	ai	NN
kim	425	6	to	to	IN
kim	425	7	Sentence	Sentence	NNP
kim	425	8	Criminals	Criminals	NNPS
kim	425	9	.	.	.
kim	426	1	That	that	DT
kim	426	2	Must	Must	MD
kim	426	3	Stop	stop	VB
kim	426	4	Now	now	RB
kim	426	5	.	.	.
kim	426	6	”	"	''
kim	426	7	Wired	wire	VBN
kim	426	8	,	,	,
kim	426	9	April	April	NNP
kim	426	10	17	17	CD
kim	426	11	,	,	,
kim	426	12	2017	2017	CD
kim	426	13	.	.	.
kim	426	14	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
kim	427	1	Tegmark	Tegmark	NNP
kim	427	2	,	,	,
kim	427	3	Max	Max	NNP
kim	427	4	.	.	.
kim	428	1	2017	2017	CD
kim	428	2	.	.	.
kim	429	1	Life	life	NN
kim	429	2	3.0	3.0	CD
kim	429	3	:	:	:
kim	429	4	Being	be	VBG
kim	429	5	Human	human	NN
kim	429	6	in	in	IN
kim	429	7	the	the	DT
kim	429	8	Age	Age	NNP
kim	429	9	of	of	IN
kim	429	10	Artificial	Artificial	NNP
kim	429	11	Intelligence	Intelligence	NNP
kim	429	12	.	.	.
kim	430	1	New	New	NNP
kim	430	2	York	York	NNP
kim	430	3	:	:	:
kim	430	4	Alfred	Alfred	NNP
kim	430	5	Knopf	Knopf	NNP
kim	430	6	.	.	.
kim	431	1	Thomson	Thomson	NNP
kim	431	2	,	,	,
kim	431	3	Judith	Judith	NNP
kim	431	4	Jarvis	Jarvis	NNP
kim	431	5	.	.	.
kim	432	1	1976	1976	CD
kim	432	2	.	.	.
kim	433	1	“	"	``
kim	433	2	Killing	kill	VBG
kim	433	3	,	,	,
kim	433	4	Letting	Letting	NNP
kim	433	5	Die	Die	NNP
kim	433	6	,	,	,
kim	433	7	and	and	CC
kim	433	8	the	the	DT
kim	433	9	Trolley	Trolley	NNP
kim	433	10	Problem	Problem	NNP
kim	433	11	.	.	.
kim	433	12	”	"	''
kim	433	13	The	the	DT
kim	433	14	Monist	Monist	NNP
kim	433	15	59	59	CD
kim	433	16	(	(	-LRB-
kim	433	17	2	2	CD
kim	433	18	)	)	-RRB-
kim	433	19	:	:	:
kim	433	20	204–17	204–17	CD
kim	433	21	.	.	.
kim	434	1	Turek	Turek	NNP
kim	434	2	,	,	,
kim	434	3	Matt	Matt	NNP
kim	434	4	.	.	.
kim	435	1	n.d	n.d	NNP
kim	435	2	.	.	.
kim	435	3	“	"	``
kim	435	4	Explainable	Explainable	NNP
kim	435	5	Artificial	Artificial	NNP
kim	435	6	Intelligence	Intelligence	NNP
kim	435	7	.	.	.
kim	435	8	”	"	''
kim	435	9	Defense	Defense	NNP
kim	435	10	Advanced	Advanced	NNP
kim	435	11	Research	Research	NNP
kim	435	12	Projects	Projects	NNPS
kim	435	13	Agency	Agency	NNP
kim	435	14	.	.	.
kim	436	1	https://www.darpa.mil/program/explainable-artificial-intelligence	https://www.darpa.mil/program/explainable-artificial-intelligence	NNP
kim	436	2	.	.	.
kim	437	1	Vallor	Vallor	NNP
kim	437	2	,	,	,
kim	437	3	Shannon	Shannon	NNP
kim	437	4	.	.	.
kim	438	1	2015	2015	CD
kim	438	2	.	.	.
kim	439	1	“	"	``
kim	439	2	Moral	Moral	NNP
kim	439	3	Deskilling	Deskilling	NNP
kim	439	4	and	and	CC
kim	439	5	Upskilling	upskille	VBG
kim	439	6	in	in	IN
kim	439	7	a	a	DT
kim	439	8	New	New	NNP
kim	439	9	Machine	Machine	NNP
kim	439	10	Age	Age	NNP
kim	439	11	:	:	:
kim	439	12	Reflections	reflection	NNS
kim	439	13	on	on	IN
kim	439	14	the	the	DT
kim	439	15	Ambiguous	ambiguous	JJ
kim	439	16	Future	Future	NNP
kim	439	17	of	of	IN
kim	439	18	Character	Character	NNP
kim	439	19	.	.	.
kim	439	20	”	"	''
kim	439	21	Philosophy	Philosophy	NNP
kim	439	22	&	&	CC
kim	439	23	Technology	Technology	NNP
kim	439	24	28	28	CD
kim	439	25	(	(	-LRB-
kim	439	26	1	1	CD
kim	439	27	)	)	-RRB-
kim	439	28	:	:	:
kim	439	29	107–24	107–24	CD
kim	439	30	.	.	.
kim	440	1	https://doi.org/10.1007/s13347-014-0156-9	https://doi.org/10.1007/s13347-014-0156-9	CD
kim	440	2	.	.	.
kim	441	1	Wallach	Wallach	NNP
kim	441	2	,	,	,
kim	441	3	Wendell	Wendell	NNP
kim	441	4	.	.	.
kim	442	1	2009	2009	CD
kim	442	2	.	.	.
kim	443	1	Moral	Moral	NNP
kim	443	2	Machines	Machines	NNPS
kim	443	3	Teaching	Teaching	NNP
kim	443	4	Robots	Robots	NNP
kim	443	5	Right	right	RB
kim	443	6	from	from	IN
kim	443	7	Wrong	wrong	NN
kim	443	8	.	.	.
kim	444	1	Oxford	Oxford	NNP
kim	444	2	:	:	:
kim	444	3	University	University	NNP
kim	444	4	Press	Press	NNP
kim	444	5	.	.	.
kim	445	1	Welch	Welch	NNP
kim	445	2	,	,	,
kim	445	3	Chris	Chris	NNP
kim	445	4	.	.	.
kim	446	1	2018	2018	CD
kim	446	2	.	.	.
kim	447	1	“	"	``
kim	447	2	Google	Google	NNP
kim	447	3	Just	just	RB
kim	447	4	Gave	give	VBD
kim	447	5	a	a	DT
kim	447	6	Stunning	Stunning	NNP
kim	447	7	Demo	Demo	NNP
kim	447	8	of	of	IN
kim	447	9	Assistant	Assistant	NNP
kim	447	10	Making	make	VBG
kim	447	11	an	an	DT
kim	447	12	Actual	actual	JJ
kim	447	13	Phone	Phone	NNP
kim	447	14	Call	call	NN
kim	447	15	.	.	.
kim	447	16	”	"	''
kim	447	17	The	the	DT
kim	447	18	Verge	Verge	NNP
kim	447	19	.	.	.
kim	448	1	May	May	NNP
kim	448	2	8	8	CD
kim	448	3	,	,	,
kim	448	4	2018	2018	CD
kim	448	5	.	.	.
kim	448	6	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
kim	448	7	.	.	.