The jobs we'll lose to machines -- and the ones we won't | Anthony Goldbloom

604,026 views ・ 2016-08-31

TED


μ•„λž˜ μ˜λ¬Έμžλ§‰μ„ λ”λΈ”ν΄λ¦­ν•˜μ‹œλ©΄ μ˜μƒμ΄ μž¬μƒλ©λ‹ˆλ‹€.

λ²ˆμ—­: Kwangmin Lee κ²€ν† : Ju Hye Lim
00:12
So this is my niece.
0
12968
1262
이 μ•„μ΄λŠ” 제 μ‘°μΉ΄μž…λ‹ˆλ‹€.
00:14
Her name is Yahli.
1
14644
1535
이름은 얄리이고
00:16
She is nine months old.
2
16203
1511
9κ°œμ›” 밖에 μ•ˆ λμ–΄μš”.
00:18
Her mum is a doctor, and her dad is a lawyer.
3
18201
2528
μ–„λ¦¬μ˜ μ—„λ§ˆλŠ” μ˜μ‚¬μ΄κ³ , μ•„λΉ λŠ” λ²•μ‘°μΈμž…λ‹ˆλ‹€.
00:21
By the time Yahli goes to college,
4
21269
2006
얄리가 λŒ€ν•™μƒμ΄ 될 λ•Œμ―€μ΄λ©΄
00:23
the jobs her parents do are going to look dramatically different.
5
23299
3253
μ–„λ¦¬μ˜ λΆ€λͺ¨λ‹˜μ˜ 직업은 μ§€κΈˆκ³ΌλŠ” ν™•μ—°νžˆ λ‹€λ₯Ό κ²λ‹ˆλ‹€.
00:27
In 2013, researchers at Oxford University did a study on the future of work.
6
27347
5073
2013년에 μ˜₯μŠ€ν¬λ“œ λŒ€ν•™μ—μ„œλŠ” μ§μ—…μ˜ λ―Έλž˜μ— κ΄€ν•΄ μ—°κ΅¬ν–ˆμŠ΅λ‹ˆλ‹€.
00:32
They concluded that almost one in every two jobs have a high risk
7
32766
4139
ν˜„μ‘΄ν•˜λŠ” μ§μ—…μ˜ 절반이 λ―Έλž˜μ—λŠ” 기계듀에 μ˜ν•΄ λŒ€μ²΄λ  ν™•λ₯ μ΄ λ†’λ‹€λŠ”
00:36
of being automated by machines.
8
36929
1824
결둠이 λ‚¬μŠ΅λ‹ˆλ‹€.
00:40
Machine learning is the technology
9
40388
1905
기계 ν•™μŠ΅μ΄ 이 λ³€ν™”μ˜ λŒ€λΆ€λΆ„μ˜ μ›μΈμž…λ‹ˆλ‹€.
00:42
that's responsible for most of this disruption.
10
42317
2278
00:44
It's the most powerful branch of artificial intelligence.
11
44619
2790
인곡지λŠ₯ λΆ„μ•Όμ—μ„œ κ°€μž₯ κ°•λ ₯ν•œ λΆ„μ•Όμž…λ‹ˆλ‹€.
00:47
It allows machines to learn from data
12
47433
1882
기계 ν•™μŠ΅μ€ 기계듀이 데이터λ₯Ό 톡해 ν•™μŠ΅ν•˜κ³ 
00:49
and mimic some of the things that humans can do.
13
49339
2592
인간이 ν•  수 μžˆλŠ” 일을 일뢀 λ”°λΌν•˜λŠ” 것을 κ°€λŠ₯μΌ€ ν•©λ‹ˆλ‹€.
00:51
My company, Kaggle, operates on the cutting edge of machine learning.
14
51955
3415
제 νšŒμ‚¬ 카글은 기계 ν•™μŠ΅ 개발의 μ΅œμ²¨λ‹¨μ„ 달리고 μžˆμŠ΅λ‹ˆλ‹€.
00:55
We bring together hundreds of thousands of experts
15
55394
2386
μ €ν¬λŠ” 수천 λͺ…이 λ„˜λŠ” 전문가듀을 λͺ¨μ•„μ„œ
00:57
to solve important problems for industry and academia.
16
57804
3118
학계와 산업이 직면해 μžˆλŠ” 문제λ₯Ό ν•΄κ²°ν•˜λ„λ‘ ν•©λ‹ˆλ‹€.
01:01
This gives us a unique perspective on what machines can do,
17
61279
3222
이 일은 μ €ν¬μ—κ²Œ 기계듀이 ν•  수 μžˆλŠ” 것과
01:04
what they can't do
18
64525
1235
ν•  수 μ—†λŠ” 것
01:05
and what jobs they might automate or threaten.
19
65784
2939
그리고 μ–΄λ–€ 직업이 기계에 μ˜ν•΄ λŒ€μ²΄λ μ§€ μ•Œ 수 있게 ν•΄μ€λ‹ˆλ‹€.
01:09
Machine learning started making its way into industry in the early '90s.
20
69316
3550
기계 ν•™μŠ΅μ€ 1990λ…„λŒ€ μ΄ˆκΈ°μ— λ“±μž₯ν–ˆμŠ΅λ‹ˆλ‹€.
01:12
It started with relatively simple tasks.
21
72890
2124
μ΄ˆλ°˜μ—λŠ” μƒλŒ€μ μœΌλ‘œ κ°„λ‹¨ν•œ 일만 ν–ˆμŠ΅λ‹ˆλ‹€.
01:15
It started with things like assessing credit risk from loan applications,
22
75406
4115
μ˜ˆμ»¨λŒ€ λŒ€μΆœ μ‹ μ²­μžλ“€μ˜ μ‹ μš© 등급을 ν‰κ°€ν•˜κ±°λ‚˜
01:19
sorting the mail by reading handwritten characters from zip codes.
23
79545
4053
νŽΈμ§€ λ΄‰νˆ¬μ— μ†κΈ€μ”¨λ‘œ 적힌 우편번호λ₯Ό μ½λŠ” μ •λ„μ˜€μ£ .
01:24
Over the past few years, we have made dramatic breakthroughs.
24
84036
3169
μ§€λ‚œ λͺ‡ λ…„ λ™μ•ˆ 기계 ν•™μŠ΅ λΆ„μ•Όμ—μ„œλŠ” 획기적인 λ°œμ „μ„ μ΄λ£¨μ–΄λƒˆμŠ΅λ‹ˆλ‹€.
01:27
Machine learning is now capable of far, far more complex tasks.
25
87586
3916
κ·Έ κ²°κ³Ό, 기계 ν•™μŠ΅μ„ 톡해 훨씬 더 λ³΅μž‘ν•œ 일을 ν•  수 있게 λ˜μ—ˆμŠ΅λ‹ˆλ‹€.
01:31
In 2012, Kaggle challenged its community
26
91860
3231
2012λ…„μ—λŠ” 카글이 기계 ν•™μŠ΅μ„ 톡해
01:35
to build an algorithm that could grade high-school essays.
27
95115
3189
고등학ꡐ 에세이λ₯Ό 채점할 수 μžˆλŠ” μ•Œκ³ λ¦¬μ¦˜μ„ λ§Œλ“œλŠ” λŒ€νšŒλ₯Ό μ—΄μ—ˆλŠ”λ°
01:38
The winning algorithms were able to match the grades
28
98328
2604
μš°μŠΉν•œ μ•Œκ³ λ¦¬μ¦˜μ΄ 맀긴 μ μˆ˜λŠ” μ‹€μ œ μ„ μƒλ‹˜λ“€μ΄
01:40
given by human teachers.
29
100956
1665
맀긴 μ μˆ˜μ™€ μΌμΉ˜ν–ˆμŠ΅λ‹ˆλ‹€.
01:43
Last year, we issued an even more difficult challenge.
30
103092
2984
μž‘λ…„μ—λŠ” 더 μ–΄λ €μš΄ 문제λ₯Ό λƒˆμŠ΅λ‹ˆλ‹€.
01:46
Can you take images of the eye and diagnose an eye disease
31
106100
2953
눈의 μ‚¬μ§„λ§Œμ„ 가지고 당뇨병성 망막증을
01:49
called diabetic retinopathy?
32
109077
1694
μ§„λ‹¨ν•˜λŠ” κ²ƒμ΄μ—ˆμ£ .
01:51
Again, the winning algorithms were able to match the diagnoses
33
111164
4040
λŒ€νšŒμ—μ„œ μš°μŠΉν•œ μ•Œκ³ λ¦¬μ¦˜μ€ μ•ˆκ³Ό μ˜μ‚¬λ“€μ˜ 진단과
01:55
given by human ophthalmologists.
34
115228
1825
또 같은 κ²°κ³Όλ₯Ό λƒˆμŠ΅λ‹ˆλ‹€.
01:57
Now, given the right data, machines are going to outperform humans
35
117561
3212
μ˜¬λ°”λ₯Έ λ°μ΄ν„°λ§Œ 주어진닀면 κΈ°κ³„λŠ” 인간보닀 이런 μž‘μ—…μ„
02:00
at tasks like this.
36
120797
1165
훨씬 더 μž˜ν•˜κ²Œ 될 κ²λ‹ˆλ‹€.
02:01
A teacher might read 10,000 essays over a 40-year career.
37
121986
3992
μ„ μƒλ‹˜ ν•œ λͺ…은 40년에 걸쳐 만 개의 에세이λ₯Ό 읽고
02:06
An ophthalmologist might see 50,000 eyes.
38
126407
2360
μ•ˆκ³Ό μ˜μ‚¬ ν•œ λͺ…은 5만 개의 λˆˆμ„ μ§„λ£Œν•  수 있겠죠.
02:08
A machine can read millions of essays or see millions of eyes
39
128791
3913
ν•˜μ§€λ§Œ κΈ°κ³„λŠ” κ³ μž‘ λͺ‡ λΆ„ μ•ˆμ— 수만 개의 눈과 에세이λ₯Ό
02:12
within minutes.
40
132728
1276
μ§„λ£Œν•˜κ³  읽을 수 μžˆμŠ΅λ‹ˆλ‹€.
02:14
We have no chance of competing against machines
41
134456
2858
인간은 μ΄λ ‡κ²Œ 반볡적이고 λ°©λŒ€ν•œ μ–‘μ˜ μž‘μ—…μ„ ν•˜λŠ” 데 μžˆμ–΄μ„œ
02:17
on frequent, high-volume tasks.
42
137338
2321
기계λ₯Ό λ›°μ–΄λ„˜μ„ 수 μ—†μŠ΅λ‹ˆλ‹€.
02:20
But there are things we can do that machines can't do.
43
140665
3724
ν•˜μ§€λ§Œ μΈκ°„λ§Œμ΄ ν•  수 μžˆλŠ” 일도 μžˆμŠ΅λ‹ˆλ‹€.
02:24
Where machines have made very little progress
44
144791
2200
기계듀은 μƒˆλ‘œμš΄ 상황에 λŒ€μ²˜ν•˜λŠ” 법을
02:27
is in tackling novel situations.
45
147015
1854
아직 λ°°μš°μ§€ λͺ»ν–ˆμŠ΅λ‹ˆλ‹€.
02:28
They can't handle things they haven't seen many times before.
46
148893
3899
기계듀은 μ „λ‘€κ°€ μ—†λŠ” 일은 μ²˜λ¦¬ν•˜μ§€ λͺ»ν•©λ‹ˆλ‹€.
02:33
The fundamental limitations of machine learning
47
153321
2584
기계 ν•™μŠ΅μ˜ 본질적인 ν•œκ³„λŠ”
02:35
is that it needs to learn from large volumes of past data.
48
155929
3394
λ°©λŒ€ν•œ μ–‘μ˜ μ „λ‘€λ₯Ό 톡해 ν•™μŠ΅ν•΄μ•Ό ν•œλ‹€λŠ” κ²ƒμž…λ‹ˆλ‹€.
02:39
Now, humans don't.
49
159347
1754
인간듀은 그럴 ν•„μš”κ°€ μ—†μ£ .
02:41
We have the ability to connect seemingly disparate threads
50
161125
3030
μš°λ¦¬λŠ” 관련이 μ—†μ–΄ λ³΄μ΄λŠ” μ£Όμ œλ“€μ„ μ΄μ–΄μ„œ
02:44
to solve problems we've never seen before.
51
164179
2238
전에 본적 μ—†λŠ” 문제λ₯Ό ν’€ 수 μžˆμŠ΅λ‹ˆλ‹€
02:46
Percy Spencer was a physicist working on radar during World War II,
52
166808
4411
νΌμ‹œ μŠ€νŽœμ„œλŠ” 제2μ°¨ 세계 λŒ€μ „ λ•Œ λ ˆμ΄λ”λ₯Ό μ—°κ΅¬ν•œ λ¬Όλ¦¬ν•™μžμ˜€μŠ΅λ‹ˆλ‹€.
02:51
when he noticed the magnetron was melting his chocolate bar.
53
171243
3013
κ·ΈλŠ” λ§ˆκ·Έλ„€νŠΈλ‘ μ΄ μ΄ˆμ½œλ¦Ώμ„ λ…Ήμ΄λŠ” 것을 λ°œκ²¬ν•˜κ³ λŠ”
02:54
He was able to connect his understanding of electromagnetic radiation
54
174970
3295
μ „μžκΈ° 방사선에 λŒ€ν•œ 지식과
02:58
with his knowledge of cooking
55
178289
1484
μš”λ¦¬μ— λŒ€ν•œ 지식을 μ—°κ²°ν•΄μ„œ
02:59
in order to invent -- any guesses? -- the microwave oven.
56
179797
3258
μ•„μ‹œλŠ” 뢄이 μžˆμœΌμ‹ μ§€ λͺ¨λ₯΄κ² μ§€λ§Œ μ „μžλ ˆμΈμ§€λ₯Ό 발λͺ…ν–ˆμŠ΅λ‹ˆλ‹€.
03:03
Now, this is a particularly remarkable example of creativity.
57
183444
3073
νŠΉνžˆλ‚˜ 더 λ†€λΌμš΄ 창의λ ₯ 발휘의 μ‚¬λ‘€μž…λ‹ˆλ‹€.
03:06
But this sort of cross-pollination happens for each of us in small ways
58
186541
3664
ν•˜μ§€λ§Œ 이런 혼재적 μ°½μ˜μ„±μ„ ν•„μš”λ‘œ ν•˜λŠ” 상황은 μš°λ¦¬μ—κ²Œ
03:10
thousands of times per day.
59
190229
1828
ν•˜λ£¨μ— μžμž˜ν•˜κ²Œ λͺ‡μ²œ λ²ˆμ”© μΌμ–΄λ‚©λ‹ˆλ‹€.
03:12
Machines cannot compete with us
60
192501
1661
κΈ°κ³„λŠ” μƒˆλ‘œμš΄ 상황에 λŒ€μ²˜ν•˜λŠ” κ²ƒμ—μ„œλŠ”
03:14
when it comes to tackling novel situations,
61
194186
2251
인간을 이길 수 μ—†μŠ΅λ‹ˆλ‹€.
03:16
and this puts a fundamental limit on the human tasks
62
196461
3117
이둜 인해 기계에 μ˜ν•΄ μžλ™ν™” 될 μΈκ°„μ˜ μž‘μ—…μ—
03:19
that machines will automate.
63
199602
1717
근본적인 ν•œκ³„κ°€ μ‘΄μž¬ν•©λ‹ˆλ‹€.
03:22
So what does this mean for the future of work?
64
202041
2405
κ·Έλ ‡λ‹€λ©΄ 이것이 μ§μ—…μ˜ λ―Έλž˜μ— λŒ€ν•΄ 뭘 μ•”μ‹œν• κΉŒμš”?
03:24
The future state of any single job lies in the answer to a single question:
65
204804
4532
μ–΄λ–€ 직업이든, κ·Έκ²ƒμ˜ λ―Έλž˜λŠ” ν•œ 가지에 μ˜ν•΄ κ²°μ •λ©λ‹ˆλ‹€.
03:29
To what extent is that job reducible to frequent, high-volume tasks,
66
209360
4981
이 μž‘μ—…μ΄ 반볡적이고 λ°©λŒ€ν•œ μ–‘μœΌλ‘œ μΆ•μ†Œλ  수 μžˆλŠ”μ§€
03:34
and to what extent does it involve tackling novel situations?
67
214365
3253
그리고 μƒˆλ‘œμš΄ 상황에 μ–Όλ§ˆλ‚˜ 많이 λŒ€μ²˜ν•΄μ•Ό ν•˜λŠ”μ§€ μž…λ‹ˆλ‹€.
03:37
On frequent, high-volume tasks, machines are getting smarter and smarter.
68
217975
4035
반볡적이고 λ°©λŒ€ν•œ μ–‘μ˜ μž‘μ—…μ—μ„œλŠ” 기계듀이 더 λ˜‘λ˜‘ν•΄μ§€κ³  μžˆμŠ΅λ‹ˆλ‹€.
03:42
Today they grade essays. They diagnose certain diseases.
69
222034
2714
μ˜€λŠ˜λ‚ μ—λŠ” 기계가 에세이도 μ±„μ ν•˜κ³  μ§ˆλ³‘λ„ μ§„λ‹¨ν•˜λ‹ˆκΉŒμš”.
03:44
Over coming years, they're going to conduct our audits,
70
224772
3157
λ―Έλž˜μ—λŠ” 기계가 νšŒκ³„ 감사도 맑고
03:47
and they're going to read boilerplate from legal contracts.
71
227953
2967
법적 계약을 맺을 λ•Œμ˜ ν‘œμ€€ ν•©μ˜μ•ˆλ„ κ²€ν† ν•  κ²ƒμž…λ‹ˆλ‹€.
03:50
Accountants and lawyers are still needed.
72
230944
1997
λ¬Όλ‘  νšŒκ³„μ‚¬μ™€ λ³€ν˜Έμ‚¬λŠ” μ—¬μ „νžˆ ν•„μš”ν•˜κ² μ£ .
03:52
They're going to be needed for complex tax structuring,
73
232965
2682
λ³΅μž‘ν•œ μ‘°μ„Έ μ—…λ¬΄λ‚˜ μ „λ‘€ μ—†λŠ” μ†Œμ†‘μ„ ν•˜λŠ” 데 ν•„μš”ν•˜κ² μ£ .
03:55
for pathbreaking litigation.
74
235671
1357
ν•˜μ§€λ§Œ κΈ°κ³„λŠ” κ·Έλ“€μ˜ μž…μ§€λ₯Ό λΉΌμ•—κ³ 
03:57
But machines will shrink their ranks
75
237052
1717
03:58
and make these jobs harder to come by.
76
238793
1872
일자리 κ΅¬ν•˜λŠ” 것을 더 νž˜λ“€κ²Œ λ§Œλ“€ κ²λ‹ˆλ‹€.
04:00
Now, as mentioned,
77
240689
1151
μ œκ°€ λ§ν–ˆλ“―μ΄
04:01
machines are not making progress on novel situations.
78
241864
2949
κΈ°κ³„λŠ” μƒˆλ‘œμš΄ 상황에 ν•œν•΄μ„œλŠ” 진전을 보이지 μ•ŠμŠ΅λ‹ˆλ‹€.
04:04
The copy behind a marketing campaign needs to grab consumers' attention.
79
244837
3457
λ§ˆμΌ€νŒ… 캠페인의 λ¬Έκ΅¬λŠ” μ†ŒλΉ„μžμ˜ 관심을 μ‚¬λ‘œμž‘μ•„μ•Ό ν•˜κ³ 
04:08
It has to stand out from the crowd.
80
248318
1715
λŒ€μ€‘λ“€ μ†μ—μ„œ 돋보여야 ν•©λ‹ˆλ‹€.
04:10
Business strategy means finding gaps in the market,
81
250057
2444
경영 μ „λž΅μ„ μ§œλŠ” 것은 아무도 ν•˜κ³  μžˆμ§€ μ•Šμ€
04:12
things that nobody else is doing.
82
252525
1756
ν‹ˆμƒˆ μ‹œμž₯을 μ°ΎλŠ” κ²ƒμž…λ‹ˆλ‹€.
04:14
It will be humans that are creating the copy behind our marketing campaigns,
83
254305
4118
κ²°κ΅­μ—λŠ” λ§ˆμΌ€νŒ… 캠페인 λ’€μ—μ„œ 문ꡬλ₯Ό λ§Œλ“œλŠ” μ£Όμ²΄λŠ” 인간이고
04:18
and it will be humans that are developing our business strategy.
84
258447
3517
경영 μ „λž΅μ„ μ§œλŠ” 것도 인간일 κ²ƒμž…λ‹ˆλ‹€.
04:21
So Yahli, whatever you decide to do,
85
261988
2817
κ·ΈλŸ¬λ‹ˆκΉŒ 얄리야, λ„€κ°€ 무슨 κΏˆμ„ 가지든
04:24
let every day bring you a new challenge.
86
264829
2361
맀일 맀일이 μƒˆλ‘œμš΄ 도전을 κ°€μ Έλ‹€ μ£ΌκΈΈ λ°”λž€λ‹€.
04:27
If it does, then you will stay ahead of the machines.
87
267587
2809
κ·Έλ ‡κ²Œ 되면 λ„€κ°€ 기계보닀 μ•žμ„œ μžˆμ„ 수 μžˆμ„ν…Œλ‹ˆκΉŒ.
04:31
Thank you.
88
271126
1176
κ°μ‚¬ν•©λ‹ˆλ‹€.
04:32
(Applause)
89
272326
3104
(λ°•μˆ˜)
이 μ›Ήμ‚¬μ΄νŠΈ 정보

이 μ‚¬μ΄νŠΈλŠ” μ˜μ–΄ ν•™μŠ΅μ— μœ μš©ν•œ YouTube λ™μ˜μƒμ„ μ†Œκ°œν•©λ‹ˆλ‹€. μ „ 세계 졜고의 μ„ μƒλ‹˜λ“€μ΄ κ°€λ₯΄μΉ˜λŠ” μ˜μ–΄ μˆ˜μ—…μ„ 보게 될 κ²ƒμž…λ‹ˆλ‹€. 각 λ™μ˜μƒ νŽ˜μ΄μ§€μ— ν‘œμ‹œλ˜λŠ” μ˜μ–΄ μžλ§‰μ„ 더블 ν΄λ¦­ν•˜λ©΄ κ·Έκ³³μ—μ„œ λ™μ˜μƒμ΄ μž¬μƒλ©λ‹ˆλ‹€. λΉ„λ””μ˜€ μž¬μƒμ— 맞좰 μžλ§‰μ΄ μŠ€ν¬λ‘€λ©λ‹ˆλ‹€. μ˜κ²¬μ΄λ‚˜ μš”μ²­μ΄ μžˆλŠ” 경우 이 문의 양식을 μ‚¬μš©ν•˜μ—¬ λ¬Έμ˜ν•˜μ‹­μ‹œμ˜€.

https://forms.gle/WvT1wiN1qDtmnspy7