????人們每天都會(huì)基于對(duì)風(fēng)險(xiǎn)的判斷做出各種決定。比如，在路上沒(méi)車(chē)的時(shí)候，急著趕公交車(chē)的你可能會(huì)橫穿馬路。但如果馬路上的車(chē)很多，你就不會(huì)這么做了。在某些危急時(shí)刻，人們必須在一瞬間做出決定。比如，正開(kāi)車(chē)時(shí)，有個(gè)小孩突然跑到汽車(chē)前面，而你的左右兩方又有其它危險(xiǎn)——比如一邊是只貓，另一邊是懸崖峭壁，這時(shí)你要如何抉擇？你會(huì)冒著自己的生命危險(xiǎn)去保護(hù)其他人的安全嗎？

????不過(guò)，面對(duì)這些道德問(wèn)題時(shí)，自動(dòng)駕駛汽車(chē)技術(shù)卻沒(méi)有快速或是確定的方法，甚至可以說(shuō)，毫無(wú)辦法。汽車(chē)廠(chǎng)商們都面臨著算法帶來(lái)的道德困境。車(chē)載電腦已經(jīng)可以幫我們泊車(chē)和自動(dòng)巡航，甚至可以在面臨重大安全問(wèn)題的時(shí)刻控制車(chē)輛。但這也意味著，自動(dòng)駕駛技術(shù)將面臨人類(lèi)有時(shí)會(huì)遭遇的艱難抉擇。

????那么，應(yīng)該如何給電腦編制“道德算法”呢？

????·計(jì)算每一種可能結(jié)果，選取傷亡數(shù)字最低的那一條路走。每個(gè)生命都要被平等對(duì)待。

????·計(jì)算每一種可能結(jié)果，選取兒童傷亡數(shù)字最低的那一條路走。

????·賦值計(jì)算：人命20分，貓命4分，狗命2分，馬命1分。然后計(jì)算每種規(guī)避措施造成后果的總分值，選取總分最低的那一條路走。那么，一大群狗的得分要超過(guò)兩只貓，車(chē)子就會(huì)向另一個(gè)方向規(guī)避，救下這一群狗。

????可是，如果車(chē)載電腦將車(chē)?yán)锏鸟{駛員和乘客也計(jì)算在內(nèi)呢？要是車(chē)外的生命總分要比車(chē)?yán)锏母咴趺崔k？要是人們知道，在必要情況下，車(chē)子會(huì)按照既定程序犧牲他們，他們還愿意坐進(jìn)車(chē)?yán)飭幔?/p>

????法國(guó)圖盧茲經(jīng)濟(jì)學(xué)院的讓·弗朗西斯·博納豐最近發(fā)表的一篇研究認(rèn)為，這些問(wèn)題沒(méi)有正確或錯(cuò)誤的答案。在此次調(diào)查中，博納豐選取了亞馬遜公司土耳其機(jī)器人項(xiàng)目（Mechanical Turk）的數(shù)百名員工作為訪(fǎng)問(wèn)對(duì)象。調(diào)查人員首先問(wèn)道：如果為了拯救一位或更多行人，車(chē)子就得轉(zhuǎn)向并撞上障礙物，從而導(dǎo)致駕駛員喪生，那么是否還應(yīng)該進(jìn)行這樣的規(guī)避？隨后調(diào)查人員會(huì)增減被拯救的行人數(shù)量。博納豐發(fā)現(xiàn)，大多數(shù)人都在原則上同意對(duì)汽車(chē)進(jìn)行編程以盡量減少傷亡人數(shù)，但談到這些情景的具體細(xì)節(jié)時(shí)，他們就不那么確定了。他們支持其他人使用自動(dòng)駕駛汽車(chē)，自己卻不那么熱心。人們往往都有實(shí)利主義本能，認(rèn)為應(yīng)該犧牲車(chē)內(nèi)人員的生命來(lái)救助更多的車(chē)外人員——除非這些車(chē)內(nèi)人員就是他們自己。

????智能機(jī)器

????科幻小說(shuō)家們總是愛(ài)寫(xiě)關(guān)于機(jī)器人暴政的故事，比如機(jī)器人占領(lǐng)了世界（如《終結(jié)者》電影以及其它許多文藝作品），每個(gè)人說(shuō)的話(huà)都會(huì)被監(jiān)聽(tīng)和分析（就像奧威爾的小說(shuō)《1984》里描寫(xiě)的一樣）等等。這些情景暫時(shí)不可能發(fā)生，但科幻小說(shuō)中的許多奇思妙想正在成為主流科技?；ヂ?lián)網(wǎng)和云計(jì)算提供的平臺(tái)已經(jīng)造就了很多技術(shù)飛躍，顯示了人工智能相對(duì)于人類(lèi)的巨大優(yōu)勢(shì)。

????在斯坦利·庫(kù)布里克的經(jīng)典電影《2001漫游太空》中，我們已經(jīng)可以見(jiàn)到一絲未來(lái)的影子。在電影中，電腦已經(jīng)可以根據(jù)任務(wù)的優(yōu)先性做出決定。飛船上的電腦HAL說(shuō)道：“這個(gè)任務(wù)對(duì)我太重要了，我不能讓你危害到它。”如今，從手機(jī)到汽車(chē)，機(jī)器智能已經(jīng)在我們身邊的許多設(shè)備上出現(xiàn)了。據(jù)英特爾公司預(yù)測(cè)，到2020年，全球?qū)⒂?.52億輛聯(lián)網(wǎng)汽車(chē)行駛在路面上，它們每年將產(chǎn)生多達(dá)1.1億億字節(jié)的數(shù)據(jù)，足以裝滿(mǎn)4萬(wàn)多個(gè)250 GB的硬盤(pán)。那么它們有多智能呢？用英特爾公司的話(huà)說(shuō)，幾乎和你一樣智能。屆時(shí)，聯(lián)網(wǎng)汽車(chē)將分享和分析一系列數(shù)據(jù)，從而在行駛中隨時(shí)做出決策。在大多數(shù)情況下，無(wú)人駕駛汽車(chē)的確比有人駕駛的汽車(chē)更安全，但我們所擔(dān)心的正是那些異常情況。

????科幻小說(shuō)家阿西莫夫著名的“機(jī)器人三定律”，為我們提出了未來(lái)智能設(shè)備如何在危急情況下進(jìn)行決策的有益建議。

????1、機(jī)器人不能傷害人，也不能不作為而坐視人受到傷害。

????2、機(jī)器人必須服從人類(lèi)的命令，除非人類(lèi)的命令違背第一原則。

????3、在不違背第一及第二原則的情況下，機(jī)器人必須保護(hù)自己。

????阿西莫夫甚至還在“三定律”之前增加了一條級(jí)別高于三定律的“第零定律”：

????·機(jī)器人不能傷害人類(lèi)社會(huì)，也不能不作為而坐視人類(lèi)社會(huì)受到傷害。

????阿西莫多雖然沒(méi)能幫我們解決“撞車(chē)”悖論，不過(guò)隨著傳感器技術(shù)的進(jìn)一步發(fā)展，數(shù)據(jù)的來(lái)源越來(lái)越多，數(shù)據(jù)處理能力越來(lái)越強(qiáng)，“撞車(chē)”決策已經(jīng)被簡(jiǎn)化為冷冰冰的數(shù)據(jù)分析。

????當(dāng)然，軟件愛(ài)鬧Bug是出了名的。如果有人惡意篡改這些系統(tǒng)，會(huì)造成怎樣的災(zāi)難？到了機(jī)器智能真的能從人類(lèi)手中接管方向盤(pán)的那一天，又會(huì)發(fā)生什么？這樣做是否正確？到時(shí)候，購(gòu)車(chē)者能否購(gòu)買(mǎi)可以編程的“道德配置”，對(duì)自己的車(chē)子進(jìn)行“道德定制”？既然有的汽車(chē)上帖著“我不為任何人踩剎車(chē)”的保險(xiǎn)杠車(chē)貼，到時(shí)候會(huì)不會(huì)有“我不為任何人踩剎車(chē)”的人工智能？如果是這樣的話(huà)，你怎么才能知道車(chē)子在危機(jī)時(shí)刻會(huì)做出怎樣的反應(yīng)——就算你知道，你又是否愿意坐上這樣一臺(tái)車(chē)呢？

????此外還有法律上的問(wèn)題。如果一輛車(chē)本可以采取措施減少傷亡，但它卻沒(méi)有這樣做，那會(huì)怎樣？如果它根據(jù)“道德計(jì)算”，直接從行人身上碾過(guò)去了怎么辦？這些都是人類(lèi)駕駛汽車(chē)時(shí)所要擔(dān)負(fù)的責(zé)任，但機(jī)器是按指令行事的，那么應(yīng)該由誰(shuí)（或者什么東西）來(lái)承擔(dān)責(zé)任？如今智能手機(jī)、機(jī)場(chǎng)監(jiān)控設(shè)備甚至連Facebook的面部識(shí)別技術(shù)都在不斷進(jìn)步，對(duì)于計(jì)算機(jī)來(lái)說(shuō)，識(shí)別物體并且根據(jù)車(chē)速和路況迅速計(jì)算出一系列可能后果，并選擇和采取相應(yīng)行動(dòng)，已經(jīng)不是很困難的事了。到那個(gè)時(shí)候，置身事中的你，很可能連選擇的機(jī)會(huì)都沒(méi)有了。（財(cái)富中文網(wǎng)）

????本文作者Bill Buchanan是愛(ài)丁堡龍比亞大學(xué)分布式計(jì)算及網(wǎng)絡(luò)與安全中心主任，本文最初發(fā)表于《The Conversation》。

????譯者：樸成奎

????審校：任文科

????We make decisions every day based on risk – perhaps running across a road to catch a bus if the road is quiet, but not if it’s busy. Sometimes these decisions must be made in an instant, in the face of dire circumstances: a child runs out in front of your car, but there are other dangers to either side, say a cat and a cliff. How do you decide? Do you risk your own safety to protect that of others?

????Now that self-driving cars are here and with no quick or sure way of overriding the controls – or even none at all – car manufacturers are faced with an algorithmic ethical dilemma. On-board computers in cars are already parking for us, driving on cruise control, and could take control in safety-critical situations. But that means they will be faced with the difficult choices that sometimes face humans.

????How to program a computer’s ethical calculus?

????? Calculate the lowest number of injuries for each possible outcome, and take that route. Every living instance would be treated the same.

????? Calculate the lowest number of injuries for children for each possible outcome, and take that route.

????? Allocate values of 20 for each human, four for a cat, two for a dog, and one for a horse. Then calculate the total score for each in the impact, and take the route with the lowest score. So a big group of dogs would rank more highly than two cats, and the car would react to save the dogs.

????What if the car also included its driver and passengers in this assessment, with the implication that sometimes those outside the car would score more highly than those within it? Who would willingly climb aboard a car programmed to sacrifice them if needs be?

????A recent study by Jean-Francois Bonnefon from the Toulouse School of Economics in France suggested that there’s no right or wrong answer to these questions. The research used several hundred workers found through Amazon’s Mechanical Turk to analyze viewpoints on whether one or more pedestrians could be saved when a car swerves and hits a barrier, killing the driver. Then they varied the number of pedestrians who could be saved. Bonnefon found that most people agreed with the principle of programming cars to minimize death toll, but when it came to the exact details of the scenarios they were less certain. They were keen for others to use self-driving cars, but less keen themselves. So people often feel a utilitarian instinct to save the lives of others and sacrifice the car’s occupant, except when that occupant is them.

????Intelligent machines

????Science fiction writers have had plenty of leash to write about robots taking over the world (Terminator and many others), or where everything that’s said is recorded and analyzed (such as in Orwell’s 1984). It’s taken a while to reach this point, but many staples of science fiction are in the process of becoming mainstream science and technology. The internet and cloud computing have provided the platform upon which quantum leaps of progress are made, showcasing artificial intelligence against the human.

????In Stanley Kubrick’s seminal film 2001: A Space Odyssey, we see hints of a future, where computers make decisions on the priorities of their mission, with the ship’s computer HAL saying: “This mission is too important for me to allow you to jeopardize it”. Machine intelligence is appearing in our devices, from phones to cars. Intel predicts that there will be 152 million connected cars by 2020, generating over 11 petabytes of data every year – enough to fill more than 40,000 250 GB hard disks. How intelligent? As Intel puts it, (almost) as smart as you. Cars will share and analyze a range data in order to make decisions on the move. It’s true enough that in most cases driverless cars are likely to be safer than humans, but it’s the outliers that we’re concerned with.

????The author Isaac Asimov’s famous three laws of robotics proposed how future devices will cope with the need to make decisions in dangerous circumstances.

????? A robot may not injure a human being or, through inaction, allow a human being to come to harm.

????? A robot must obey the orders given it by human beings, except where such orders would conflict with the First Law.

????? A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.

????He even added a more fundamental “0th law” preceding the others:

????? A robot may not harm humanity, or, by inaction, allow humanity to come to harm.

????Asimov did not tackle our ethical dilemma of the car crash, but with greater sensors to gather data, more sources of data to draw from, and greater processing power, the decision to act is reduced to a cold act of data analysis.

????Of course software is notoriously buggy. What havoc could malicious actors who compromise these systems wreak? And what happens at the point that machine intelligence takes control from the human? Will it be right to do so? Could a future buyer purchase programmable ethical options with which to customize their car? The artificial intelligence equivalent of a bumper sticker that says “I break for nobody”? In which case, how would you know how cars were likely to act – and would you climb aboard if you did?

????Then there are the legal issues. What if a car could have intervened to save lives but didn’t? Or if it ran people down deliberately based on its ethical calculus? This is the responsibility we bear as humans when we drive a car, but machines follow orders, so who (or what) carries the responsibility for a decision? As we see with improving face recognition in smartphones, airport monitors and even on Facebook, it’s not too difficult for a computer to identify objects, quickly calculate the consequences based on car speed and road conditions in order to calculate a set of outcomes, pick one, and act. And when it does so, it’s unlikely you’ll have an choice in the matter.

????Bill Buchanan is head of the center for distributed computing, networks and security at Edinburgh Napier University. This article originally appeared on The Conversation.