TED2019
Janelle Shane: The danger of AI is weirder than you think
Janelle Shane: Bahaya kecerdasan buatan lebih aneh dari dugaanmu
Filmed:
Readability: 4.2
376,501 views
Menurut peneliti kecerdasan buatan (AI) Janelle Shane, bahaya AI bukanlah pemberontakan, tapi AI mungkin benar-benar melakukan perintah kita. Dengan membahas keanehan, juga perilaku unik algoritme AI dalam upaya memecahkan masalah manusia, seperti menciptakan rasa es krim baru atau mengenali mobil di jalan, Shane menunjukkan alasan AI belum sebanding dengan otak manusia.
Janelle Shane - AI researcher
While moonlighting as a research scientist, Janelle Shane found fame documenting the often hilarious antics of AI algorithms. Full bio
While moonlighting as a research scientist, Janelle Shane found fame documenting the often hilarious antics of AI algorithms. Full bio
Double-click the English transcript below to play the video.
00:13
So, artificial intelligence
0
1765
3000
Kecerdasan buatan (AI)
00:16
is known for disrupting
all kinds of industries.
all kinds of industries.
1
4789
3529
dikenal mengganggu
berbagai macam industri.
berbagai macam industri.
00:20
What about ice cream?
2
8961
2043
Bagaimana dengan es krim?
00:23
What kind of mind-blowing
new flavors could we generate
new flavors could we generate
3
11879
3639
Rasa baru mencengangkan seperti apa
yang bisa kita ciptakan
yang bisa kita ciptakan
00:27
with the power of an advanced
artificial intelligence?
artificial intelligence?
4
15542
2976
dengan kecanggihan kecerdasan buatan?
00:31
So I teamed up with a group of coders
from Kealing Middle School
from Kealing Middle School
5
19011
4161
Saya bekerja sama dengan pembuat kode
dari SMP Kealing
dari SMP Kealing
00:35
to find out the answer to this question.
6
23196
2241
untuk mencari jawaban pertanyaan ini.
00:37
They collected over 1,600
existing ice cream flavors,
existing ice cream flavors,
7
25461
5081
Mereka mengumpulkan 1.600 lebih
rasa es krim yang sudah ada,
rasa es krim yang sudah ada,
00:42
and together, we fed them to an algorithm
to see what it would generate.
to see what it would generate.
8
30566
5522
lalu kami memasukkannya ke algoritme
untuk melihat hasilnya.
untuk melihat hasilnya.
00:48
And here are some of the flavors
that the AI came up with.
that the AI came up with.
9
36112
3753
Ini beberapa rasa yang diciptakan AI.
00:52
[Pumpkin Trash Break]
10
40444
1471
[Patahan Sampah Labu]
00:53
(Laughter)
11
41939
1402
(Tertawa)
00:55
[Peanut Butter Slime]
12
43365
2469
[Lendir Selai Kacang]
00:58
[Strawberry Cream Disease]
13
46822
1343
[Penyakit Es Krim Stroberi]
01:00
(Laughter)
14
48189
2126
(Tertawa)
01:02
These flavors are not delicious,
as we might have hoped they would be.
as we might have hoped they would be.
15
50339
4597
Rasa-rasa ini tidak seenak harapan kami.
01:06
So the question is: What happened?
16
54960
1864
Pertanyaannya, apa yang terjadi?
01:08
What went wrong?
17
56848
1394
Apa yang salah?
01:10
Is the AI trying to kill us?
18
58266
1959
Apakah AI ingin membunuh kita?
01:13
Or is it trying to do what we asked,
and there was a problem?
and there was a problem?
19
61027
4310
Atau hanya menuruti perintah,
tapi ada masalah?
tapi ada masalah?
01:18
In movies, when something
goes wrong with AI,
goes wrong with AI,
20
66567
2464
Dalam film, jika AI bermasalah,
01:21
it's usually because the AI has decided
21
69055
2712
itu biasanya karena kecerdasan buatan itu
01:23
that it doesn't want to obey
the humans anymore,
the humans anymore,
22
71791
2272
tidak ingin patuh pada manusia lagi.
Dia memiliki tujuan sendiri,
terima kasih.
terima kasih.
01:26
and it's got its own goals,
thank you very much.
thank you very much.
23
74087
2623
01:29
In real life, though,
the AI that we actually have
the AI that we actually have
24
77266
3216
Namun di kehidupan nyata,
AI yang kita miliki
AI yang kita miliki
01:32
is not nearly smart enough for that.
25
80506
1863
tidak cukup pintar untuk itu.
01:34
It has the approximate computing power
26
82781
2982
AI yang ada memiliki daya komputasi
01:37
of an earthworm,
27
85787
1276
seperti cacing tanah
01:39
or maybe at most a single honeybee,
28
87087
3403
atau setidaknya seperti lebah madu.
01:42
and actually, probably maybe less.
29
90514
2215
Mungkin kurang dari itu.
01:44
Like, we're constantly learning
new things about brains
new things about brains
30
92753
2594
Kita terus mempelajari hal baru
tentang otak
tentang otak
01:47
that make it clear how much our AIs
don't measure up to real brains.
don't measure up to real brains.
31
95371
4360
dan menunjukkan bahwa AI
tak sebanding dengan otak kita.
tak sebanding dengan otak kita.
01:51
So today's AI can do a task
like identify a pedestrian in a picture,
like identify a pedestrian in a picture,
32
99755
5663
AI saat ini
bisa mengenali pejalan kaki dalam foto,
bisa mengenali pejalan kaki dalam foto,
01:57
but it doesn't have a concept
of what the pedestrian is
of what the pedestrian is
33
105442
2983
tapi tidak paham apa itu pejalan kaki
02:00
beyond that it's a collection
of lines and textures and things.
of lines and textures and things.
34
108449
4824
selain bahwa itu hanya kumpulan garis,
tekstur, dan lainnya.
tekstur, dan lainnya.
02:05
It doesn't know what a human actually is.
35
113792
2521
AI tak tahu apa itu manusia.
02:08
So will today's AI
do what we ask it to do?
do what we ask it to do?
36
116822
3282
Apakah AI saat ini
akan menuruti perintah kita?
akan menuruti perintah kita?
02:12
It will if it can,
37
120128
1594
Ya, jika bisa.
02:13
but it might not do what we actually want.
38
121746
2726
Namun, mungkin tak sesuai keinginan kita.
02:16
So let's say that you
were trying to get an AI
were trying to get an AI
39
124496
2415
Anggap Anda ingin kecerdasan buatan
02:18
to take this collection of robot parts
40
126935
2619
mengambil suku cadang robot,
02:21
and assemble them into some kind of robot
to get from Point A to Point B.
to get from Point A to Point B.
41
129578
4197
menyusunnya menjadi robot,
dan pindah dari Poin A ke Poin B.
dan pindah dari Poin A ke Poin B.
02:25
Now, if you were going to try
and solve this problem
and solve this problem
42
133799
2481
Jika Anda ingin selesaikan masalah ini
dengan membuat program komputer biasa,
02:28
by writing a traditional-style
computer program,
computer program,
43
136304
2351
02:30
you would give the program
step-by-step instructions
step-by-step instructions
44
138679
3417
Anda memberi instruksi
langkah demi langkah pada program
langkah demi langkah pada program
untuk mengambil suku cadang,
02:34
on how to take these parts,
45
142120
1329
02:35
how to assemble them
into a robot with legs
into a robot with legs
46
143473
2407
menyusunnya menjadi robot berkaki,
02:37
and then how to use those legs
to walk to Point B.
to walk to Point B.
47
145904
2942
dan cara menggunakan kakinya
untuk berjalan ke Poin B.
untuk berjalan ke Poin B.
02:41
But when you're using AI
to solve the problem,
to solve the problem,
48
149441
2340
Namun, jika memakai kecerdasan buatan,
02:43
it goes differently.
49
151805
1174
caranya berbeda.
02:45
You don't tell it
how to solve the problem,
how to solve the problem,
50
153003
2382
Tak perlu menjelaskan
cara menyelesaikan masalah,
cara menyelesaikan masalah,
02:47
you just give it the goal,
51
155409
1479
Anda hanya memberi tujuan.
02:48
and it has to figure out for itself
via trial and error
via trial and error
52
156912
3262
AI harus mencoba mencari cara sendiri
02:52
how to reach that goal.
53
160198
1484
untuk mencapai tujuan itu.
02:54
And it turns out that the way AI tends
to solve this particular problem
to solve this particular problem
54
162254
4102
Ternyata, AI cenderung
mengatasi masalah ini
mengatasi masalah ini
02:58
is by doing this:
55
166380
1484
seperti ini:
02:59
it assembles itself into a tower
and then falls over
and then falls over
56
167888
3367
dia membentuk dirinya menjadi menara,
lalu jatuh
lalu jatuh
03:03
and lands at Point B.
57
171279
1827
dan mendarat di Poin B.
03:05
And technically, this solves the problem.
58
173130
2829
Secara teknis, masalahnya terpecahkan.
03:07
Technically, it got to Point B.
59
175983
1639
Dia sampai ke Poin B.
03:09
The danger of AI is not that
it's going to rebel against us,
it's going to rebel against us,
60
177646
4265
Bahaya kecerdasan buatan
bukan kemungkinan dia memberontak,
bukan kemungkinan dia memberontak,
03:13
it's that it's going to do
exactly what we ask it to do.
exactly what we ask it to do.
61
181935
4274
tapi AI akan benar-benar mematuhi kita.
03:18
So then the trick
of working with AI becomes:
of working with AI becomes:
62
186876
2498
Maka, trik menggunakan AI adalah
03:21
How do we set up the problem
so that it actually does what we want?
so that it actually does what we want?
63
189398
3828
bagaimana kita menyusun masalah
agar AI berfungsi sesuai harapan?
agar AI berfungsi sesuai harapan?
03:26
So this little robot here
is being controlled by an AI.
is being controlled by an AI.
64
194726
3306
Robot kecil ini dikontrol oleh AI.
03:30
The AI came up with a design
for the robot legs
for the robot legs
65
198056
2814
AI membuat desain untuk kaki robotnya
03:32
and then figured out how to use them
to get past all these obstacles.
to get past all these obstacles.
66
200894
4078
dan bisa menggunakannya
untuk melalui rintangan ini.
untuk melalui rintangan ini.
03:36
But when David Ha set up this experiment,
67
204996
2741
Namun, saat David Ha
membuat eksperimen ini,
membuat eksperimen ini,
03:39
he had to set it up
with very, very strict limits
with very, very strict limits
68
207761
2856
dia harus membuat batasan tegas
03:42
on how big the AI
was allowed to make the legs,
was allowed to make the legs,
69
210641
3292
seberapa besar kaki
yang bisa dibuat oleh AI.
yang bisa dibuat oleh AI.
03:45
because otherwise ...
70
213957
1550
Jika tidak...
03:55
(Laughter)
71
223058
3931
(Tertawa)
04:00
And technically, it got
to the end of that obstacle course.
to the end of that obstacle course.
72
228563
3745
Secara teknis,
rintangan itu berhasil dilalui.
rintangan itu berhasil dilalui.
04:04
So you see how hard it is to get AI
to do something as simple as just walk.
to do something as simple as just walk.
73
232332
4942
Sangat sulit membuat AI
melakukan hal sederhana seperti berjalan.
melakukan hal sederhana seperti berjalan.
04:09
So seeing the AI do this,
you may say, OK, no fair,
you may say, OK, no fair,
74
237298
3820
Saat melihat ini,
Anda mungkin berkata itu tidak adil.
Anda mungkin berkata itu tidak adil.
04:13
you can't just be
a tall tower and fall over,
a tall tower and fall over,
75
241142
2580
Anda tak boleh menjadi menara tinggi,
lalu jatuh.
lalu jatuh.
04:15
you have to actually, like,
use legs to walk.
use legs to walk.
76
243746
3435
Anda harus menggunakan kaki
untuk berjalan.
untuk berjalan.
04:19
And it turns out,
that doesn't always work, either.
that doesn't always work, either.
77
247205
2759
Ternyata, itu tidak selalu berhasil.
Tugas AI ini adalah bergerak cepat.
04:21
This AI's job was to move fast.
78
249988
2759
04:25
They didn't tell it that it had
to run facing forward
to run facing forward
79
253115
3593
Kami tak memintanya berlari
menghadap ke depan
menghadap ke depan
04:28
or that it couldn't use its arms.
80
256732
2258
atau tak boleh menggunakan tangan.
04:31
So this is what you get
when you train AI to move fast,
when you train AI to move fast,
81
259487
4618
Ini yang terjadi saat Anda melatih AI
bergerak cepat.
bergerak cepat.
04:36
you get things like somersaulting
and silly walks.
and silly walks.
82
264129
3534
Hasilnya adalah jungkir balik
dan cara berjalan aneh.
dan cara berjalan aneh.
Ini biasa terjadi.
04:39
It's really common.
83
267687
1400
04:41
So is twitching along the floor in a heap.
84
269667
3179
Juga bergeliat di lantai
dalam bentuk gumpalan.
dalam bentuk gumpalan.
04:44
(Laughter)
85
272870
1150
(Tertawa)
04:47
So in my opinion, you know what
should have been a whole lot weirder
should have been a whole lot weirder
86
275241
3254
Menurut pendapat saya,
yang semestinya lebih aneh
yang semestinya lebih aneh
adalah robot "Terminator".
04:50
is the "Terminator" robots.
87
278519
1396
04:52
Hacking "The Matrix" is another thing
that AI will do if you give it a chance.
that AI will do if you give it a chance.
88
280256
3755
AI mungkin akan meretas "The Matrix"
jika dicoba.
jika dicoba.
04:56
So if you train an AI in a simulation,
89
284035
2517
Jika AI dilatih dalam sebuah simulasi,
04:58
it will learn how to do things like
hack into the simulation's math errors
hack into the simulation's math errors
90
286576
4113
dia akan belajar cara meretas
kesalahan matematis simulasi itu
kesalahan matematis simulasi itu
05:02
and harvest them for energy.
91
290713
2207
dan menjadikannya energi.
Atau dia akan bisa bergerak lebih cepat
dengan bergeliat di lantai.
dengan bergeliat di lantai.
05:04
Or it will figure out how to move faster
by glitching repeatedly into the floor.
by glitching repeatedly into the floor.
92
292944
5475
Bekerja dengan AI
05:10
When you're working with AI,
93
298443
1585
tidak seperti bekerja dengan manusia,
05:12
it's less like working with another human
94
300052
2389
05:14
and a lot more like working
with some kind of weird force of nature.
with some kind of weird force of nature.
95
302465
3629
tapi seperti bekerja
dengan kekuatan alam yang aneh.
dengan kekuatan alam yang aneh.
05:18
And it's really easy to accidentally
give AI the wrong problem to solve,
give AI the wrong problem to solve,
96
306562
4623
Kesalahan sangat mudah terjadi
saat memberi perintah pada AI,
saat memberi perintah pada AI,
05:23
and often we don't realize that
until something has actually gone wrong.
until something has actually gone wrong.
97
311209
4538
dan kami sering kali tak sadar
sampai ada yang salah.
sampai ada yang salah.
05:28
So here's an experiment I did,
98
316242
2080
Ini eksperimen yang saya lakukan.
05:30
where I wanted the AI
to copy paint colors,
to copy paint colors,
99
318346
3182
Saya ingin AI menyalin warna cat
untuk menciptakan warna cat baru
05:33
to invent new paint colors,
100
321552
1746
05:35
given the list like the ones
here on the left.
here on the left.
101
323322
2987
dengan daftar seperti gambar
di sebelah kiri.
di sebelah kiri.
05:38
And here's what the AI
actually came up with.
actually came up with.
102
326798
3004
Inilah yang diciptakan oleh AI.
05:41
[Sindis Poop, Turdly, Suffer, Gray Pubic]
103
329826
3143
[Kotoran Sindi, Kerdil,
Menderita, Kemaluan Abu-Abu]
Menderita, Kemaluan Abu-Abu]
05:44
(Laughter)
104
332993
4230
(Tertawa)
Secara teknis,
05:51
So technically,
105
339177
1886
05:53
it did what I asked it to.
106
341087
1864
perintah saya dipatuhi.
05:54
I thought I was asking it for,
like, nice paint color names,
like, nice paint color names,
107
342975
3308
Saya sudah memintanya
membuat nama cat yang bagus,
membuat nama cat yang bagus,
05:58
but what I was actually asking it to do
108
346307
2307
tapi yang sebenarnya saya perintahkan
06:00
was just imitate the kinds
of letter combinations
of letter combinations
109
348638
3086
adalah meniru kombinasi huruf
06:03
that it had seen in the original.
110
351748
1905
dari daftar aslinya.
06:05
And I didn't tell it anything
about what words mean,
about what words mean,
111
353677
3098
Saya tak menjelaskan apa arti kata-kata
06:08
or that there are maybe some words
112
356799
2560
atau mungkin ada kata-kata
06:11
that it should avoid using
in these paint colors.
in these paint colors.
113
359383
2889
yang harus dihindari dalam warna cat ini.
Seluruh dunia AI ini
adalah data yang saya berikan.
adalah data yang saya berikan.
06:15
So its entire world
is the data that I gave it.
is the data that I gave it.
114
363141
3494
06:18
Like with the ice cream flavors,
it doesn't know about anything else.
it doesn't know about anything else.
115
366659
4028
Seperti rasa es krim,
AI itu tidak tahu tentang hal lain.
AI itu tidak tahu tentang hal lain.
Jadi, melalui data,
06:24
So it is through the data
116
372491
1638
06:26
that we often accidentally tell AI
to do the wrong thing.
to do the wrong thing.
117
374153
4044
kami sering salah memberi perintah
secara tak sengaja.
secara tak sengaja.
06:30
This is a fish called a tench.
118
378694
3032
Ini adalah ikan mas tinca.
06:33
And there was a group of researchers
119
381750
1815
Ada sekelompok peneliti
06:35
who trained an AI to identify
this tench in pictures.
this tench in pictures.
120
383589
3874
yang melatih AI
untuk mengenali ikan ini dalam foto.
untuk mengenali ikan ini dalam foto.
06:39
But then when they asked it
121
387487
1296
Saat mereka bertanya
bagian foto mana yang digunakan
untuk mengenali ikannya,
untuk mengenali ikannya,
06:40
what part of the picture it was actually
using to identify the fish,
using to identify the fish,
122
388807
3426
06:44
here's what it highlighted.
123
392257
1358
ini bagian yang disorot.
06:47
Yes, those are human fingers.
124
395203
2189
Ya, itu jari manusia.
06:49
Why would it be looking for human fingers
125
397416
2059
Kenapa AI mencari jari manusia
untuk mengenali ikan?
06:51
if it's trying to identify a fish?
126
399499
1921
06:54
Well, it turns out that the tench
is a trophy fish,
is a trophy fish,
127
402126
3164
Ternyata, ikan mas tinca
adalah ikan trofi.
adalah ikan trofi.
06:57
and so in a lot of pictures
that the AI had seen of this fish
that the AI had seen of this fish
128
405314
3811
Jadi, dalam banyak foto yang dilihat AI
07:01
during training,
129
409149
1151
selama pelatihan,
ikannya tampak seperti ini.
07:02
the fish looked like this.
130
410324
1490
07:03
(Laughter)
131
411838
1635
(Tertawa)
07:05
And it didn't know that the fingers
aren't part of the fish.
aren't part of the fish.
132
413497
3330
AI tak tahu
jari itu bukan bagian dari ikan.
jari itu bukan bagian dari ikan.
07:10
So you see why it is so hard
to design an AI
to design an AI
133
418808
4120
Itulah sebabnya sangat sulit
merancang AI
merancang AI
07:14
that actually can understand
what it's looking at.
what it's looking at.
134
422952
3319
yang bisa mengerti apa yang dilihatnya.
07:18
And this is why designing
the image recognition
the image recognition
135
426295
2862
Inilah alasan
merancang pengenalan gambar
merancang pengenalan gambar
07:21
in self-driving cars is so hard,
136
429181
2067
dalam mobil tanpa pengemudi
sangat sulit.
sangat sulit.
07:23
and why so many self-driving car failures
137
431272
2205
Banyak mobil tanpa pengemudi gagal,
07:25
are because the AI got confused.
138
433501
2885
karena AI menjadi bingung.
07:28
I want to talk about an example from 2016.
139
436410
4008
Saya ingin membahas contoh
dari tahun 2016.
dari tahun 2016.
07:32
There was a fatal accident when somebody
was using Tesla's autopilot AI,
was using Tesla's autopilot AI,
140
440442
4455
Terjadi kecelakaan fatal saat seseorang
mengendarai Tesla dengan autopilot.
mengendarai Tesla dengan autopilot.
07:36
but instead of using it on the highway
like it was designed for,
like it was designed for,
141
444921
3414
Namun, alih-alih memakainya di jalan tol
seperti tujuan awalnya,
seperti tujuan awalnya,
07:40
they used it on city streets.
142
448359
2205
dia menggunakannya di jalanan kota.
Yang terjadi adalah
07:43
And what happened was,
143
451239
1175
truk melaju ke depan mobil
dan mobilnya tak mengerem.
dan mobilnya tak mengerem.
07:44
a truck drove out in front of the car
and the car failed to brake.
and the car failed to brake.
144
452438
3396
07:48
Now, the AI definitely was trained
to recognize trucks in pictures.
to recognize trucks in pictures.
145
456507
4762
AI itu sudah dilatih
untuk mengenali truk dalam foto.
untuk mengenali truk dalam foto.
07:53
But what it looks like happened is
146
461293
2145
Namun, sepertinya yang terjadi adalah
07:55
the AI was trained to recognize
trucks on highway driving,
trucks on highway driving,
147
463462
2931
AI itu dilatih
untuk mengenali truk di jalan tol,
untuk mengenali truk di jalan tol,
07:58
where you would expect
to see trucks from behind.
to see trucks from behind.
148
466417
2899
di mana biasanya truk datang
dari belakang.
dari belakang.
08:01
Trucks on the side is not supposed
to happen on a highway,
to happen on a highway,
149
469340
3420
Truk di samping tidak seharusnya ada
di jalan tol.
di jalan tol.
08:04
and so when the AI saw this truck,
150
472784
3455
Saat AI melihat truk itu,
08:08
it looks like the AI recognized it
as most likely to be a road sign
as most likely to be a road sign
151
476263
4827
sepertinya AI mengenalinya
sebagai rambu lalu lintas,
sebagai rambu lalu lintas,
08:13
and therefore, safe to drive underneath.
152
481114
2273
dan melaju di bawahnya akan aman.
08:16
Here's an AI misstep
from a different field.
from a different field.
153
484114
2580
Ini kesalahan AI di bidang yang berbeda.
Baru-baru ini, Amazon menghentikan
algoritme penyortiran CV
algoritme penyortiran CV
08:18
Amazon recently had to give up
on a résumé-sorting algorithm
on a résumé-sorting algorithm
154
486718
3460
yang mereka kerjakan
08:22
that they were working on
155
490202
1220
saat mereka tahu algoritme itu
mendiskriminasi wanita.
mendiskriminasi wanita.
08:23
when they discovered that the algorithm
had learned to discriminate against women.
had learned to discriminate against women.
156
491446
3908
Itu karena mereka melatih AI
dengan contoh CV
dengan contoh CV
08:27
What happened is they had trained it
on example résumés
on example résumés
157
495378
2716
dari orang-orang
yang pernah mereka terima.
yang pernah mereka terima.
08:30
of people who they had hired in the past.
158
498118
2242
08:32
And from these examples, the AI learned
to avoid the résumés of people
to avoid the résumés of people
159
500384
4023
Dari contoh itu, AI belajar
menghindari CV orang-orang
menghindari CV orang-orang
08:36
who had gone to women's colleges
160
504431
2026
yang pernah belajar di universitas putri
08:38
or who had the word "women"
somewhere in their resume,
somewhere in their resume,
161
506481
2806
atau yang berisi kata "wanita" di CV-nya,
seperti "tim sepak bola wanita"
atau "Perhimpunan Insinyur Wanita".
atau "Perhimpunan Insinyur Wanita".
08:41
as in, "women's soccer team"
or "Society of Women Engineers."
or "Society of Women Engineers."
162
509311
4576
08:45
The AI didn't know that it wasn't supposed
to copy this particular thing
to copy this particular thing
163
513911
3974
AI itu tidak tahu
dia tak seharusnya menyalin hal tertentu
dia tak seharusnya menyalin hal tertentu
08:49
that it had seen the humans do.
164
517909
1978
dari yang dilakukan manusia.
Secara teknis,
AI itu sudah mematuhi perintah.
AI itu sudah mematuhi perintah.
08:51
And technically, it did
what they asked it to do.
what they asked it to do.
165
519911
3177
08:55
They just accidentally asked it
to do the wrong thing.
to do the wrong thing.
166
523112
2797
Mereka hanya tak sengaja
memberi perintah yang salah.
memberi perintah yang salah.
Ini sering terjadi dengan AI.
08:58
And this happens all the time with AI.
167
526653
2895
09:02
AI can be really destructive
and not know it.
and not know it.
168
530120
3591
AI bisa sangat merusak
tanpa menyadarinya.
tanpa menyadarinya.
09:05
So the AIs that recommend
new content in Facebook, in YouTube,
new content in Facebook, in YouTube,
169
533735
5078
Jadi, AI yang menyarankan konten baru
di Facebook dan YouTube
di Facebook dan YouTube
09:10
they're optimized to increase
the number of clicks and views.
the number of clicks and views.
170
538837
3539
dioptimalkan untuk meningkatkan
jumlah klik dan tontonan.
jumlah klik dan tontonan.
09:14
And unfortunately, one way
that they have found of doing this
that they have found of doing this
171
542400
3436
Sayangnya, cara yang mereka pelajari
untuk melakukan itu
untuk melakukan itu
09:17
is to recommend the content
of conspiracy theories or bigotry.
of conspiracy theories or bigotry.
172
545860
4503
adalah menyarankan konten teori konspirasi
atau kefanatikan.
atau kefanatikan.
09:22
The AIs themselves don't have any concept
of what this content actually is,
of what this content actually is,
173
550902
5302
AI itu sendiri tak mengerti
apa itu konten.
apa itu konten.
09:28
and they don't have any concept
of what the consequences might be
of what the consequences might be
174
556228
3395
AI juga tak mengerti konsekuensi
09:31
of recommending this content.
175
559647
2109
dari menyarankan konten seperti itu.
Saat kita bekerja dengan AI,
09:34
So, when we're working with AI,
176
562296
2011
09:36
it's up to us to avoid problems.
177
564331
4182
kita yang menjadi penentu
untuk menghindari masalah.
untuk menghindari masalah.
09:40
And avoiding things going wrong,
178
568537
2323
Menghindari kesalahan
mungkin mengarah pada masalah komunikasi
yang sudah ada sejak lama,
yang sudah ada sejak lama,
09:42
that may come down to
the age-old problem of communication,
the age-old problem of communication,
179
570884
4526
09:47
where we as humans have to learn
how to communicate with AI.
how to communicate with AI.
180
575434
3745
di mana manusia harus belajar
cara berkomunikasi dengan AI.
cara berkomunikasi dengan AI.
09:51
We have to learn what AI
is capable of doing and what it's not,
is capable of doing and what it's not,
181
579203
4039
Kita harus mempelajari
yang bisa dilakukan AI dan tidak.
yang bisa dilakukan AI dan tidak.
09:55
and to understand that,
with its tiny little worm brain,
with its tiny little worm brain,
182
583266
3086
Juga memahami,
dengan kemampuan berpikirnya,
dengan kemampuan berpikirnya,
09:58
AI doesn't really understand
what we're trying to ask it to do.
what we're trying to ask it to do.
183
586376
4013
AI tidak begitu mengerti perintah kita.
10:03
So in other words, we have
to be prepared to work with AI
to be prepared to work with AI
184
591148
3321
Dengan kata lain,
kita harus siap bekerja dengan AI
kita harus siap bekerja dengan AI
10:06
that's not the super-competent,
all-knowing AI of science fiction.
all-knowing AI of science fiction.
185
594493
5258
yang tak begitu kompeten dan canggih
seperti dalam fiksi ilmiah.
seperti dalam fiksi ilmiah.
10:11
We have to prepared to work with an AI
186
599775
2862
Kita harus siap bekerja dengan AI
yang kita miliki saat ini.
10:14
that's the one that we actually have
in the present day.
in the present day.
187
602661
2938
10:17
And present-day AI is plenty weird enough.
188
605623
4205
Dan AI masa kini agak aneh.
10:21
Thank you.
189
609852
1190
Terima kasih.
10:23
(Applause)
190
611066
5225
(Tepuk tangan)
ABOUT THE SPEAKER
Janelle Shane - AI researcherWhile moonlighting as a research scientist, Janelle Shane found fame documenting the often hilarious antics of AI algorithms.
Why you should listen
Janelle Shane's humor blog, AIweirdness.com, looks at, as she tells it, "the strange side of artificial intelligence." Her upcoming book, You Look Like a Thing and I Love You: How AI Works and Why It's Making the World a Weirder Place, uses cartoons and humorous pop-culture experiments to look inside the minds of the algorithms that run our world, making artificial intelligence and machine learning both accessible and entertaining.
According to Shane, she has only made a neural network-written recipe once -- and discovered that horseradish brownies are about as terrible as you might imagine.
Janelle Shane | Speaker | TED.com