Don't fear intelligent machines. Work with them | Garry Kasparov

翻译人员: Lipeng Chen 校对人员: Jing Peng

This story begins in 1985,

这个故事发生在1985年，

when at age 22,

我22岁的时候，

I became the World Chess Champion

在打败了Anatoly Karpov之后，

after beating Anatoly Karpov.

我成为了国际象棋世界冠军。

Earlier that year,

这一年的早些时候，

I played what is called simultaneous exhibition

在德国汉堡，我和32台

against 32 of the world's best chess-playing machines

世界顶尖级别的电脑进行了

in Hamburg, Germany.

一场国际象棋“车轮战”。

I won all the games,

我赢得了所有的对弈，

and then it was not considered much of a surprise

能够同时打败32台电脑，

that I could beat 32 computers at the same time.

对我来说也不算是什么大惊喜。

To me, that was the golden age.

我认为这是我的黄金时代。

(Laughter)

（笑声）

Machines were weak,

那会儿的电脑实力还弱，

and my hair was strong.

我的头发还挺茂盛。

(Laughter)

（笑声）

Just 12 years later,

就在12年后，

I was fighting for my life against just one computer

我拼劲全力与一台电脑对弈，

in a match

这场对弈，

called by the cover of "Newsweek"

《新闻周刊》的封面报道称之为

"The Brain's Last Stand."

“人脑的背水一战”。

No pressure.

不需要有压力。

(Laughter)

（笑声）

From mythology to science fiction,

从神话到科幻，

human versus machine

人类对弈电脑

has been often portrayed as a matter of life and death.

经常被形容为生死攸关的问题。

John Henry,

John Henry，

called the steel-driving man

19世纪非洲裔美国人,

in the 19th century African American folk legend,

民间传说中的"钢铁勇士“，

was pitted in a race

在开凿山石隧道时

against a steam-powered hammer

与一台蒸汽动力锤

bashing a tunnel through mountain rock.

进行了生死较量。

John Henry's legend is a part of a long historical narrative

John Henry的传奇是历史长河中

pitting humanity versus technology.

人类与科技对弈的一部分。

And this competitive rhetoric is standard now.

而现在这种对弈已是司空见惯了。

We are in a race against the machines,

我们正在与机器竞争，

in a fight or even in a war.

这是一场对抗， 更可能是一场战争。

Jobs are being killed off.

工作机会正在消失。

People are being replaced as if they had vanished from the Earth.

人们被机器所取代， 就好像他们正从地球上消失。

It's enough to think that the movies like "The Terminator" or "The Matrix"

这一切足以让我们认为 电影《终结者》和《黑客帝国》

are nonfiction.

并不是虚构。

There are very few instances of an arena

在很少的情况下才能看到

人类的身体和脑子可以与 电脑或是机器人

where the human body and mind can compete on equal terms

with a computer or a robot.

平分秋色。

Actually, I wish there were a few more.

事实上，我希望这种情况能更多些。

Instead,

然而，

it was my blessing and my curse

我成为人们仍在谈论的

to literally become the proverbial man

人类与机器的对抗中的

in the man versus machine competition

那个传奇人物，

that everybody is still talking about.

这既是我的福祉，也是对我的诅咒。

In the most famous human-machine competition since John Henry,

自从John Henry以来的 人与机器的较量中，

I played two matches

我和IBM的超级电脑

against the IBM supercomputer, Deep Blue.

深蓝（Deep Blue），对弈过两次。

Nobody remembers that I won the first match --

没人记得我赢了第一场比赛——

(Laughter)

（笑声）

(Applause)

（掌声）

In Philadelphia, before losing the rematch the following year in New York.

那是在费城。也就是 我输掉纽约那场的前一年。

But I guess that's fair.

但我觉得这也没什么。

There is no day in history, special calendar entry

在Edmund Hillary公爵和 Tenzing Norgay成功登顶珠峰前，

for all the people who failed to climb Mt. Everest

历史上从未有一天是

before Sir Edmund Hillary and Tenzing Norgay

为登顶失败的人

made it to the top.

而记载的特殊的日子。

And in 1997, I was still the world champion

在1997年会下棋的电脑问世之时，

when chess computers finally came of age.

我仍然是世界冠军。

I was Mt. Everest,

我就是那珠穆朗玛峰。

and Deep Blue reached the summit.

而深蓝成功登顶。

I should say of course, not that Deep Blue did it,

我应该说，当然了， 并不是深蓝成功了，

but its human creators --

而是创造深蓝的人——

Anantharaman, Campbell, Hoane, Hsu.

Anantharaman，Campbell Hoane，Hsu。

Hats off to them.

向他们致敬。

As always, machine's triumph was a human triumph,

一如既往，机器的胜利 就是人类的胜利，

something we tend to forget when humans are surpassed by our own creations.

当人类被自己的创造超越后， 这一点就常常被遗忘了。

Deep Blue was victorious,

深蓝取得了胜利，

but was it intelligent?

但是，是它聪明吗？

No, no it wasn't,

不，并不是，

at least not in the way Alan Turing and other founders of computer science

至少不是像Alan Turing和 其他计算科学的奠基人

had hoped.

希望的那样。

It turned out that chess could be crunched by brute force,

事实证明，一旦电脑软件运行的够快，

once hardware got fast enough

算法变得灵活起来，

and algorithms got smart enough.

象棋的下法就可以 被轻而易举地算出。

Although by the definition of the output,

尽管按照输出的定义，

grandmaster-level chess,

按大师级别来算，

Deep Blue was intelligent.

深蓝算得上是高手。

But even at the incredible speed,

但是即使以

200 million positions per second,

每秒2亿次的惊人计算速度，

Deep Blue's method

深蓝的方法

provided little of the dreamed-of insight into the mysteries of human intelligence.

也只探索了人类智慧 谜题中的一丁点儿。

Soon,

很快，

machines will be taxi drivers

机器就可以成为出租车司机、

and doctors and professors,

医生、教授，

but will they be "intelligent?"

但是他们会变得“有智慧”吗？

I would rather leave these definitions

我宁愿将这些定义

to the philosophers and to the dictionary.

留给哲学家或者是字典。

What really matters is how we humans

真正重要的是，我们人类

feel about living and working with these machines.

与电脑机器共存会是怎样的感受。

When I first met Deep Blue in 1996 in February,

当我在1996年2月 第一次遇到深蓝的时候，

I had been the world champion for more than 10 years,

我已经当了十多年的世界冠军了，

and I had played 182 world championship games

我已经参加了182场世界冠军赛了，

and hundreds of games against other top players in other competitions.

也在其他的比赛中与数不胜数的 高手们进了数百次的对弈。

I knew what to expect from my opponents

我知道我的对手会是什么样子，

and what to expect from myself.

他们也知道我的情况。

I was used to measure their moves

我习惯于通过观察他们的

and to gauge their emotional state

肢体语言和眼神，

by watching their body language and looking into their eyes.

考量他们的棋步， 摸索他们的心态。

And then I sat across the chessboard from Deep Blue.

而当我坐在深蓝的对面时，

I immediately sensed something new,

我立刻感受到了莫名的气息，

something unsettling.

有些不安。

You might experience a similar feeling

你可能有过类似的感受，

the first time you ride in a driverless car

当你第一次坐进一辆无人驾驶汽车，

or the first time your new computer manager issues an order at work.

或者是你的电计算机老板 在工作中派给你一个任务。

But when I sat at that first game,

在第一场对弈时，

I couldn't be sure

我无法确定，

what is this thing capable of.

这东西能够做什么。

Technology can advance in leaps, and IBM had invested heavily.

技术可以跨越式前进， IBM花了重金。

I lost that game.

我被打败了。

And I couldn't help wondering,

我不停地思考，

might it be invincible?

它会是无敌的吗？

Was my beloved game of chess over?

我深爱的象棋生涯完蛋了吗？

These were human doubts, human fears,

这些都是人类的怀疑和恐惧，

and the only thing I knew for sure

唯一我敢肯定的是

was that my opponent Deep Blue had no such worries at all.

我的对手深蓝 完全没有这样的烦恼。

(Laughter)

（笑声）

I fought back

在这场惨败后

after this devastating blow

我卷土重来，

to win the first match,

赢下了第一场比赛，

but the writing was on the wall.

但很明显，我注定会失败。

I eventually lost to the machine

最终还是输给了机器，

but I didn't suffer the fate of John Henry

但我的命运与John Henry大不相同，

who won but died with his hammer in his hand.

这位凿山工赢得了比赛，但却丢了性命， 死的时候手里还攥着铁锤。

[John Henry Died with a Hammer in His Hand Palmer C. Hayden]

【John Henry手握着铁锤死去 Palmer C. Hayden】

[The Museum of African American Art, Los Angeles]

【非裔美国艺术博物馆，洛杉矶】

It turned out that the world of chess

但是象棋世界

still wanted to have a human chess champion.

还是想要人类象棋冠军。

And even today,

不过即便是今天，

when a free chess app on the latest mobile phone

智能手机中的免费象棋软件

is stronger than Deep Blue,

都比深蓝还要聪明，

people are still playing chess,

人们依然在下象棋，

even more than ever before.

对它的兴趣不减反增。

Doomsayers predicted that nobody would touch the game

悲观者说没有人会再碰

that could be conquered by the machine,

可以被机器打败的游戏，

and they were wrong, proven wrong,

而他们错了，事实已经证明了，

but doomsaying has always been a popular pastime

但是一谈起科技，末日说一直是

when it comes to technology.

很流行的消遣说法。

What I learned from my own experience

我从自身的经历中所学到就是，

is that we must face our fears

如果我们要充分地利用科技，

if we want to get the most out of our technology,

就要面对我们的恐惧，

and we must conquer those fears

如果我们要最大化地激发人类的潜能，

if we want to get the best out of our humanity.

就必须战胜我们的恐惧感。

While licking my wounds,

当我自省时，

I got a lot of inspiration

我从和深蓝的对战中

from my battles against Deep Blue.

获得了很多灵感。

As the old Russian saying goes, if you can't beat them, join them.

正如古老的俄罗斯谚语所说： 如果你不能打败他们，就加入他们。

Then I thought,

后来，我就想

what if I could play with a computer --

如果我能和电脑合作，

together with a computer at my side, combining our strengths,

结合我们的优点：

human intuition plus machine's calculation,

人类的直觉和电脑的计算，

human strategy, machine tactics,

人类的策略和电脑的战术，

human experience, machine's memory.

人类的经验和电脑的记忆。

Could it be the perfect game ever played?

会不会打一场最完美的比赛呢？

My idea came to life

1998年，在一个高阶象棋赛中

in 1998 under the name of Advanced Chess

我的想法实现了，

when I played this human-plus-machine competition against another elite player.

我利用人机结合的方式跟 另一位顶尖棋手进行了对弈。

But in this first experiment,

但是在第一次试验里，

we both failed to combine human and machine skills effectively.

我们都没能有效的结合 人和机器的技能。

Advanced Chess found its home on the internet,

高阶象棋发迹于网络，

and in 2005, a so-called freestyle chess tournament

在2005年，一个被称为 自由式象棋竞标赛的比赛

produced a revelation.

诞生了。

A team of grandmasters and top machines participated,

一队顶尖高手和顶级电脑参赛了，

but the winners were not grandmasters,

但是冠军并不是顶尖高手，

not a supercomputer.

也不是超级电脑。

The winners were a pair of amateur American chess players

冠军是一对同时在操作

operating three ordinary PCs at the same time.

三台普通电脑的 业余美国棋手。

Their skill of coaching their machines

他们有效操作电脑的技能

effectively counteracted the superior chess knowledge

胜过了他们的大师级对手

of their grandmaster opponents

超高的象棋知识，

and much greater computational power of others.

也胜过了电脑更强大的计算能力。

And I reached this formulation.

我写出了这个公式。

A weak human player plus a machine

一个不强的人类棋手加上一台机器，

plus a better process is superior

加上一个更好的处理器要好过

to a very powerful machine alone,

一台非常强大的机器本身，

but more remarkably, is superior to a strong human player

但更值得注意的是， 也会超过一个强大的人类棋手

plus machine

加上机器，

and an inferior process.

和一个较差的处理器。

This convinced me that we would need

这让我确信

better interfaces to help us coach our machines

我们需要更好的界面 去帮助我们的电脑学习，

towards more useful intelligence.

来实现更有用的智能。

Human plus machine isn't the future,

人类与电脑的结合并不是将来时，

it's the present.

而是现在时。

Everybody that's used online translation

所有使用在线翻译

to get the gist of a news article from a foreign newspaper,

来了解外国报纸新闻概要的人

knowing its far from perfect.

都知道这些软件远远不够完美。

Then we use our human experience

然后我们结合我们的人类经验

to make sense out of that,

让这变得合理，

and then the machine learns from our corrections.

而电脑从我们的更改中学习升级。

This model is spreading and investing in medical diagnosis, security analysis.

这种模式广泛应用于 医疗诊断，证券分析。

The machine crunches data,

机器分析数据

calculates probabilities,

计算出概率，

gets 80 percent of the way, 90 percent,

完成了百分之八，九十的工作，

making it easier for analysis

使得分析和决策

and decision-making of the human party.

对人类来说更加简单。

But you are not going to send your kids

但是你不会用准确度90%，

to school in a self-driving car with 90 percent accuracy,

甚至99%的无人驾驶汽车

even with 99 percent.

把你的孩子送到学校去。

So we need a leap forward

所以我们需要跨越式前进，

to add a few more crucial decimal places.

提高准确度。

Twenty years after my match with Deep Blue,

在我和深蓝对弈二十年之后，

second match,

第二场比赛，

this sensational "The Brain's Last Stand" headline

这个耸人听闻的标题， “人脑的背水一战”

has become commonplace

已经变得司空见惯了，

as intelligent machines

因为智能机器

move

在每个领域，

in every sector, seemingly every day.

每一天都在工作。

But unlike in the past,

但与过去不同，

when machines replaced

那时机器替代的是

farm animals, manual labor,

农场动物，劳工，

now they are coming after people with college degrees

现在它们的目标是替代有大学学位

and political influence.

和政治影响力的人们。

And as someone who fought machines and lost,

作为曾经对抗过机器而且失败了的人，

I am here to tell you this is excellent, excellent news.

我在这里告诉你们， 这是非常非常棒的消息。

Eventually, every profession

最终，每个职业

will have to feel these pressures

都不得不感受这种压力，

or else it will mean humanity has ceased to make progress.

否则就意味着人类已经停止前进了。

We don't

技术进步的脚步

get to choose

何时何地停止，

when and where technological progress stops.

我们无从选择。

We cannot

我们无法

slow down.

放缓它的脚步。

In fact,

现实是，

we have to speed up.

我们要加速前进

Our technology excels at removing

科学技术擅长解决

difficulties and uncertainties from our lives,

我们生活中的困难和不确定性，

and so we must seek out

所以我们必须面对

ever more difficult,

更加困难，

ever more uncertain challenges.

更加不确定的挑战。

Machines have

机器可以

calculations.

计算。

We have understanding.

而我们有理解力。

Machines have instructions.

机器遵从指令。

We have purpose.

我们拥有目标。

Machines have

机器具有

objectivity.

客观性。

We have passion.

我们拥有激情。

We should not worry about what our machines can do today.

我们不应该为机器 现在可以做什么而感到担心。

Instead, we should worry about what they still cannot do today,

相反，我们应该为它们 今日仍不能做的感到担心，

because we will need the help of the new, intelligent machines

因为我们需要新的智能机器的帮助，

to turn our grandest dreams into reality.

来实现我们宏伟的梦想。

And if we fail,

如果我们失败的话，

if we fail, it's not because our machines are too intelligent,

如果我们失败， 不是因为我们的机器太聪明了，

or not intelligent enough.

或不够聪明。

If we fail, it's because we grew complacent

如果我们失败了， 是因为我们变得自满，

and limited our ambitions.

限制了我们的理想抱负。

Our humanity is not defined by any skill,

我们人类不由任何技能定义，

like swinging a hammer or even playing chess.

就像敲锤子或是下象棋。

There's one thing only a human can do.

有一件事情只有人类能做，

That's dream.

那就是梦想。

So let us dream big.

让我们的梦想更深远一些吧。

Thank you.

谢谢。

(Applause)

(掌声）