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

604,026 views ・ 2016-08-31

TED


請雙擊下方英文字幕播放視頻。

譯者: Harry Chen 審譯者: 易帆 余
00:12
So this is my niece.
0
12968
1262
這是我的姪女,
00:14
Her name is Yahli.
1
14644
1535
她的名字是雅莉,
00:16
She is nine months old.
2
16203
1511
她現在九個月大,
00:18
Her mum is a doctor, and her dad is a lawyer.
3
18201
2528
媽媽是位醫生、爸爸是位律師;
00:21
By the time Yahli goes to college,
4
21269
2006
不過等到她上大學的時候
00:23
the jobs her parents do are going to look dramatically different.
5
23299
3253
她父母親的工作將會迥然不同了。
00:27
In 2013, researchers at Oxford University did a study on the future of work.
6
27347
5073
2013年,牛津大學的研究人員 做了一個對未來工作的研究,
00:32
They concluded that almost one in every two jobs have a high risk
7
32766
4139
他們得出結論:差不多將近一半的工作
00:36
of being automated by machines.
8
36929
1824
都有被機器自動化取代的危險。
00:40
Machine learning is the technology
9
40388
1905
而「機器學習」
00:42
that's responsible for most of this disruption.
10
42317
2278
要對這種顛覆負主要的責任。
00:44
It's the most powerful branch of artificial intelligence.
11
44619
2790
它是人工智慧最呼風喚雨的分支,
00:47
It allows machines to learn from data
12
47433
1882
它讓機器得以從數據中學習,
00:49
and mimic some of the things that humans can do.
13
49339
2592
並模仿一些人類可以做到的事情。
00:51
My company, Kaggle, operates on the cutting edge of machine learning.
14
51955
3415
我的公司「Kaggle」算是能操控 機器學習的尖端科技公司。
00:55
We bring together hundreds of thousands of experts
15
55394
2386
我們召集了成千上萬的專家
00:57
to solve important problems for industry and academia.
16
57804
3118
為產、學界解決重要的難題。
01:01
This gives us a unique perspective on what machines can do,
17
61279
3222
所以我們可以從獨特的角度 來觀察機器可以做什麽、
01:04
what they can't do
18
64525
1235
不可以做什麽。
01:05
and what jobs they might automate or threaten.
19
65784
2939
哪些工作可以被自動化或者受到威脅。
01:09
Machine learning started making its way into industry in the early '90s.
20
69316
3550
機器學習是在90年代初 進入產業界的,
01:12
It started with relatively simple tasks.
21
72890
2124
一開始,它只是執行一些 簡單的任務。
01:15
It started with things like assessing credit risk from loan applications,
22
75406
4115
像評估貸款申請的信用風險、
01:19
sorting the mail by reading handwritten characters from zip codes.
23
79545
4053
查看郵遞區號的手寫字碼 來分類郵件。
01:24
Over the past few years, we have made dramatic breakthroughs.
24
84036
3169
過去幾年來我們已經做出 多項重大的突破,
01:27
Machine learning is now capable of far, far more complex tasks.
25
87586
3916
機器學習現在已經可以完成 非常覆雜的任務。
01:31
In 2012, Kaggle challenged its community
26
91860
3231
在 2012 年 Kaggle 給自家社群出了一道難題,
01:35
to build an algorithm that could grade high-school essays.
27
95115
3189
要大家設計出一個演算法 來評判高中作文。
01:38
The winning algorithms were able to match the grades
28
98328
2604
獲勝的演算法給出的分數居然 和真正老師給出的分數相符
01:40
given by human teachers.
29
100956
1665
去年,我們出了一道更難的題目:
01:43
Last year, we issued an even more difficult challenge.
30
103092
2984
01:46
Can you take images of the eye and diagnose an eye disease
31
106100
2953
你可不可以藉由眼球的影像
診斷出一種叫「糖尿病視網膜病變」的眼疾?
01:49
called diabetic retinopathy?
32
109077
1694
01:51
Again, the winning algorithms were able to match the diagnoses
33
111164
4040
果然,獲勝的演算法給出的診斷
可以和人類眼科醫師的診斷相媲美。
01:55
given by human ophthalmologists.
34
115228
1825
01:57
Now, given the right data, machines are going to outperform humans
35
117561
3212
只要給定正確的數據 , 機器在類似的任務中
02:00
at tasks like this.
36
120797
1165
將完全超越人類。
02:01
A teacher might read 10,000 essays over a 40-year career.
37
121986
3992
一位老師,在他的40年職業生涯中 也許只能審閱10000篇作文
02:06
An ophthalmologist might see 50,000 eyes.
38
126407
2360
一名眼科醫生,大概可以看50,000隻眼睛
02:08
A machine can read millions of essays or see millions of eyes
39
128791
3913
但一部機器可以在短短幾分鐘內 讀完上百萬篇文章
02:12
within minutes.
40
132728
1276
或是看完上百萬顆眼睛。
02:14
We have no chance of competing against machines
41
134456
2858
在頻繁、大批量的任務上 我們無法與機器抗衡。
02:17
on frequent, high-volume tasks.
42
137338
2321
02:20
But there are things we can do that machines can't do.
43
140665
3724
不過還是有我們能做 而機器做不到的事情,
02:24
Where machines have made very little progress
44
144791
2200
機器在解決複雜的新情況方面
02:27
is in tackling novel situations.
45
147015
1854
進展甚微。
02:28
They can't handle things they haven't seen many times before.
46
148893
3899
它們對還沒看到很多次的事情無法掌握。
02:33
The fundamental limitations of machine learning
47
153321
2584
機器學習的先天限制就是:
02:35
is that it needs to learn from large volumes of past data.
48
155929
3394
它需要從大量的過往資料中學習。
02:39
Now, humans don't.
49
159347
1754
人類就不一樣了,
02:41
We have the ability to connect seemingly disparate threads
50
161125
3030
我們有一種能把看似毫不相關的事物 聯系起來的能力,
02:44
to solve problems we've never seen before.
51
164179
2238
從而解決我們先前還不曾見過的難題。
02:46
Percy Spencer was a physicist working on radar during World War II,
52
166808
4411
波西‧史賓塞是二次世界大戰期間, 從事雷達研究的物理學家,
02:51
when he noticed the magnetron was melting his chocolate bar.
53
171243
3013
當他注意到磁控管不斷融化 他的巧克力棒時,
02:54
He was able to connect his understanding of electromagnetic radiation
54
174970
3295
他能夠把他對電磁波的認知
02:58
with his knowledge of cooking
55
178289
1484
與烹飪的知識做結合,
02:59
in order to invent -- any guesses? -- the microwave oven.
56
179797
3258
因此發明了--各位猜猜是什麽? 微波爐。
03:03
Now, this is a particularly remarkable example of creativity.
57
183444
3073
這是個特別傑出的創新例子
03:06
But this sort of cross-pollination happens for each of us in small ways
58
186541
3664
但是這種跨領域的碰撞,
每天在我們的周遭會上演好幾千回。
03:10
thousands of times per day.
59
190229
1828
03:12
Machines cannot compete with us
60
192501
1661
在解決新的棘手問題方面 機器無法與我們媲美,
03:14
when it comes to tackling novel situations,
61
194186
2251
03:16
and this puts a fundamental limit on the human tasks
62
196461
3117
而這使機器自動化取代人工
03:19
that machines will automate.
63
199602
1717
受到了限制。
03:22
So what does this mean for the future of work?
64
202041
2405
那麽這對未來的工作意味著什麽呢?
03:24
The future state of any single job lies in the answer to a single question:
65
204804
4532
未來工作的狀態完全取決於一個問題:
03:29
To what extent is that job reducible to frequent, high-volume tasks,
66
209360
4981
「該工作有多少程度可以縮減成 經常性、高產量的任務,
03:34
and to what extent does it involve tackling novel situations?
67
214365
3253
以及有多少程度是在解決新的棘手問題?」
03:37
On frequent, high-volume tasks, machines are getting smarter and smarter.
68
217975
4035
對於那些頻繁,大批量的任務, 機器變得越來越聰明。
03:42
Today they grade essays. They diagnose certain diseases.
69
222034
2714
今天它們能給作文打分數、 診斷特定疾病,
03:44
Over coming years, they're going to conduct our audits,
70
224772
3157
過了幾年後,它們將可以進行審計、
03:47
and they're going to read boilerplate from legal contracts.
71
227953
2967
從法律合約中解讀法律語言。
03:50
Accountants and lawyers are still needed.
72
230944
1997
盡管會計師和律師還是需要的
03:52
They're going to be needed for complex tax structuring,
73
232965
2682
但僅能研究覆雜的稅務結構及 無例可循的法律問題,
03:55
for pathbreaking litigation.
74
235671
1357
03:57
But machines will shrink their ranks
75
237052
1717
不過機器將會減少他們的就業機會, 增加就業難度。
03:58
and make these jobs harder to come by.
76
238793
1872
04:00
Now, as mentioned,
77
240689
1151
如同我說過的:
04:01
machines are not making progress on novel situations.
78
241864
2949
機器在處理複雜的新情境上 沒有進步!
04:04
The copy behind a marketing campaign needs to grab consumers' attention.
79
244837
3457
行銷推案的文宣必須擄獲消費者的青睞,
04:08
It has to stand out from the crowd.
80
248318
1715
它必須脫俗出眾。
04:10
Business strategy means finding gaps in the market,
81
250057
2444
商業策略必須在市場上找到一些
04:12
things that nobody else is doing.
82
252525
1756
其它人還沒開始做的領域。
04:14
It will be humans that are creating the copy behind our marketing campaigns,
83
254305
4118
人類才是營銷文案的創造者,
04:18
and it will be humans that are developing our business strategy.
84
258447
3517
人類才是商業戰略的拓展人
04:21
So Yahli, whatever you decide to do,
85
261988
2817
所以,雅莉,不管妳決定要做什麼,
04:24
let every day bring you a new challenge.
86
264829
2361
讓每一天帶給妳新的挑戰,
04:27
If it does, then you will stay ahead of the machines.
87
267587
2809
如果是這樣,那麼妳將永遠領先機器一步。
04:31
Thank you.
88
271126
1176
謝謝大家!
04:32
(Applause)
89
272326
3104
(掌聲)
關於本網站

本網站將向您介紹對學習英語有用的 YouTube 視頻。 您將看到來自世界各地的一流教師教授的英語課程。 雙擊每個視頻頁面上顯示的英文字幕,從那裡播放視頻。 字幕與視頻播放同步滾動。 如果您有任何意見或要求,請使用此聯繫表與我們聯繫。

https://forms.gle/WvT1wiN1qDtmnspy7