声明: 本站全部内容源自互联网,不进行任何盈利行为
仅做 整合 / 美化 处理
So this is my niece.
这是我的侄女。
Her name is Yahli.
她叫Yahl。
She is nine months old.
她只有九个月大。
Her mum is a doctor, and her dad is a lawyer.
她妈妈是一名医生， 爸爸是一名律师。
By the time Yahli goes to college,
等到Yahli上大学的时候，
the jobs her parents do are going to look dramatically different.
像她父母这样的工作将面目全非。
In 2013, researchers at Oxford University did a study on the future of work.
2013年，牛津大学的研究人员 做了一项关于未来就业的研究。
They concluded that almost one in every two jobs have a high risk
他们得出结论：差不多将近 一半的工作都有被机器
of being automated by machines.
自动化取代的危险。
Machine learning is the technology
而机器学习
that's responsible for most of this disruption.
应对这种颠覆负主要责任。
It's the most powerful branch of artificial intelligence.
它是人工智能最强大的分支。
It allows machines to learn from data
允许机器从现有数据中学习，
and mimic some of the things that humans can do.
并模仿人类的所作所为。
My company, Kaggle, operates on the cutting edge of machine learning.
我的公司Kaggle 专注于尖端的机器学习。
We bring together hundreds of thousands of experts
我们召集了成千上万的专家
to solve important problems for industry and academia.
正为工业和学术界 寻找重要问题的答案。
This gives us a unique perspective on what machines can do,
因此，我们可以从独特的视角来观察，
what they can't do
机器可以做什么，不可以做什么，
and what jobs they might automate or threaten.
哪些工作可以被自动化或受到威胁。
Machine learning started making its way into industry in the early '90s.
机器学习是在90年代初 进入人们的视野。
It started with relatively simple tasks.
一开始，它只是执行 一些相对简单的任务。
It started with things like assessing credit risk from loan applications,
像评估贷款申请的信用风险，
sorting the mail by reading handwritten characters from zip codes.
通过识别手写的邮政编码来检索邮件。
Over the past few years, we have made dramatic breakthroughs.
在过去几年里，我们取得了突破性进展。
Machine learning is now capable of far, far more complex tasks.
现在，机器学习可以 完成非常复杂的任务。
In 2012, Kaggle challenged its community
2012年，Kaggle给当地学校出了个难题，
to build an algorithm that could grade high-school essays.
设计一个算法来评判高中作文。
The winning algorithms were able to match the grades
获胜的算法给出的分数居然
given by human teachers.
和真正老师给出的分数相符。
Last year, we issued an even more difficult challenge.
去年，我们出了一道更难的题。
Can you take images of the eye and diagnose an eye disease
你能从拍摄出的眼睛图像中 诊断出糖尿病性
called diabetic retinopathy?
视网膜病变吗？
Again, the winning algorithms were able to match the diagnoses
再一次，获胜的演算法给出的诊断
given by human ophthalmologists.
和眼科医生的诊断相符。
Now, given the right data, machines are going to outperform humans
类似于这样的任务， 只要给定正确的数据，
at tasks like this.
机器将完全超越人类。
A teacher might read 10,000 essays over a 40-year career.
一位老师在40年的职业生涯中 可能审阅一万篇作文。
An ophthalmologist might see 50,000 eyes.
一名眼科医生，大概可以检查 5万只眼睛。
A machine can read millions of essays or see millions of eyes
但在短短几分钟之内， 机器可以审阅百万篇文章
within minutes.
或检查数百万只眼睛。
We have no chance of competing against machines
对于频繁，大批量的任务
on frequent, high-volume tasks.
我们无法与机器抗衡。
But there are things we can do that machines can't do.
但有些事情机器却无能为力。
Where machines have made very little progress
机器在解决新情况方面
is in tackling novel situations.
进展甚微。
They can't handle things they haven't seen many times before.
它们还不能处理未曾反复接触的事情。
The fundamental limitations of machine learning
机器学习致命的局限性在于
is that it needs to learn from large volumes of past data.
它需要从大量已知的数据中总结经验。
Now, humans don't.
人类则不然。
We have the ability to connect seemingly disparate threads
我们有一种能把看似毫不相关的事物 联系起来的能力，
to solve problems we've never seen before.
从而解决从未见过的问题
Percy Spencer was a physicist working on radar during World War II,
Percy Spencer是一个物理学家， 在二战期间从事雷达的研究工作，
when he noticed the magnetron was melting his chocolate bar.
他注意到磁控管融化了他的巧克力。
He was able to connect his understanding of electromagnetic radiation
他从对电磁辐射的理解
with his knowledge of cooking
联想到烹饪，
in order to invent -- any guesses? -- the microwave oven.
因此发明了——猜猜是什么？—— 微波炉。
Now, this is a particularly remarkable example of creativity.
这是个非常杰出的创新例子。
But this sort of cross-pollination happens for each of us in small ways
但这种跨界转型，每天正以 难以察觉的方式在我们身边
thousands of times per day.
发生成千上百次。
Machines cannot compete with us
在创新方面
when it comes to tackling novel situations,
机器无法与我们抗衡。
and this puts a fundamental limit on the human tasks
这将使机器自动化取代人工
that machines will automate.
受到限制。
So what does this mean for the future of work?
那么这对未来的工作意味着什么呢？
The future state of any single job lies in the answer to a single question:
未来工作的状态 完全取决于一个问题：
To what extent is that job reducible to frequent, high-volume tasks,
这种工作在多大程度上可以简化为 频繁，大批量的任务，
and to what extent does it involve tackling novel situations?
又涉及多少对创新能力的要求？
On frequent, high-volume tasks, machines are getting smarter and smarter.
对于那些频繁，大批量的任务， 机器变得越来越智能。
Today they grade essays. They diagnose certain diseases.
如今， 它们可以评判作文， 诊断某些疾病。
Over coming years, they're going to conduct our audits,
再过几年，它们将可以进行审计，
and they're going to read boilerplate from legal contracts.
将能审阅法律合同样本。
Accountants and lawyers are still needed.
尽管会计师和律师还是需要的。
They're going to be needed for complex tax structuring,
但他们只需要研究复杂的税收结构，
for pathbreaking litigation.
或无先例的诉讼过程。
But machines will shrink their ranks
但机器将会挤占他们的位置，
and make these jobs harder to come by.
增加就业难度。
Now, as mentioned,
如上所述，
machines are not making progress on novel situations.
在创新方面机器没有取得太大进展。
The copy behind a marketing campaign needs to grab consumers' attention.
营销文案需要抓住消费者的心理。
It has to stand out from the crowd.
脱颖而出是关键。
Business strategy means finding gaps in the market,
商业策略需要找到市场上
things that nobody else is doing.
还无人问津的空白。
It will be humans that are creating the copy behind our marketing campaigns,
人类将是营销文案的创造者，
and it will be humans that are developing our business strategy.
人类才能推动商业战略发展。
So Yahli, whatever you decide to do,
所以Yahli，无论你将来决定做什么，
let every day bring you a new challenge.
让每一天都带给你新的挑战。
If it does, then you will stay ahead of the machines.
如果是那样， 你的未来将无法被机器取代。
Thank you.
谢谢。
(Applause)
（掌声 ）