嘗試精讀新方法

通讀全文標(biāo)出不懂得單詞和表達(dá)

查字典記錄上面標(biāo)出的表達(dá)的釋義

進(jìn)行段落翻譯并與參考翻譯進(jìn)行對(duì)比（嘗試翻譯為斜體字，在參考翻譯中將關(guān)鍵翻譯短語(yǔ)黑體標(biāo)出）

根據(jù)上面兩段中文釋義轉(zhuǎn)翻為英文，關(guān)注關(guān)鍵短語(yǔ)

分析文章結(jié)構(gòu)和大意

記錄自己的觀點(diǎn)和看法

harness ? n ?馬具；牽索；日常工作 ? ?v 套上馬具；利用（自然力）產(chǎn)生能源

merge ? ?合并

think-tank ? ?智囊團(tuán)

consultancy ? ?顧問(wèn)工作

reckon ? ?認(rèn)為；計(jì)算

grandios ? ?浮夸的，不切實(shí)際的

fret ? ?苦惱，煩惱

vibartion ? ?震動(dòng)

nudge ? ?輕推

surveillance ? ?監(jiān)視

idle ? ?虛度，浪費(fèi)時(shí)間；使...空轉(zhuǎn)

ubiquitous ? ?無(wú)所不在的

unguardely ? ?不小心的，謹(jǐn)慎的

consent ? ?同意，贊許

預(yù)測(cè)：forecast predict

公司：company firm

購(gòu)買：acquisition purchasing

職工：staff worker employee

檢測(cè)：surveillance????screen for ? ?track ? ?monitor

隨著人工智能推動(dòng)科技工業(yè)進(jìn)步，工作會(huì)變得更加公平——或者更加壓抑。

人工智能向科技行業(yè)以外擴(kuò)張，它可能會(huì)讓工作場(chǎng)所變得更公平，也可能變得更高壓

As artificial intelligence pushes beyond the tech industries, work could become fairer —— or more oppressive.

人工智能正慢慢走進(jìn)商業(yè)。就像我們這周的報(bào)告所揭示的，所有類型的公司都利用人工智能來(lái)預(yù)測(cè)需求，解雇員工和處理客戶。2017年，這些公司在人工智能相關(guān)的兼并和收購(gòu)上花費(fèi)了近22億元，大約是2015年的26倍。麥肯錫全球研究所，這個(gè)做顧問(wèn)工作的智囊團(tuán)認(rèn)為只把AI運(yùn)用到銷售，供給鏈能夠在接下來(lái)的20年創(chuàng)造價(jià)值2.7trn美元經(jīng)濟(jì)價(jià)值，包括利潤(rùn)和效率。谷歌老板一直聲稱AI帶給人類的將會(huì)比火和電更多。

人工智能橫沖直撞，闖入了商業(yè)領(lǐng)域。各種各樣的公司都在利用人工智能來(lái)預(yù)測(cè)需求、雇傭員工、與客戶打交道。2017年。企業(yè)在AI方面的并購(gòu)支出達(dá)220億美元上下，大約是2015年的26倍。咨詢公司麥肯錫的內(nèi)部智庫(kù)麥肯錫全球研究所認(rèn)為，僅僅是將人工智能應(yīng)用到營(yíng)銷、銷售和供應(yīng)鏈上，未來(lái)20年就能創(chuàng)造2.7萬(wàn)億美元的經(jīng)濟(jì)價(jià)值，包括利潤(rùn)和效率。谷歌的老板甚至宣稱對(duì)人類而言，人工智能比火和電的用處更大。

AI is barging its way to business. companie of all types areharnessing AI to predict demand, hire workers and deal with customers. In 2017, firms spend about $22bn on AI-related merges and purchasing acquisition, about 26 times more than in 2016. McKinsey Global Institute, a think-tank within a consultancy, reckons that only applying AI to marking, sales and supply chains will creat economic value, including profits and efficiencies, in 20 years?of $2.7trn over next 20 years. Google's boss has gone so far as to declare that AI will do more for human?humanity than fire and electric.

如此不切實(shí)際的預(yù)測(cè)點(diǎn)燃了希望也引發(fā)了焦慮。很多人為AI對(duì)工作的破壞大于創(chuàng)造它感到焦慮。擁有和產(chǎn)生數(shù)據(jù)的障礙可能會(huì)導(dǎo)致在每個(gè)產(chǎn)業(yè)產(chǎn)生大量的統(tǒng)治的公司。

如此宏偉的預(yù)測(cè)既點(diǎn)亮了希望，也帶來(lái)了焦慮。許多人擔(dān)心AI摧毀就業(yè)的速度會(huì)比創(chuàng)造就業(yè)機(jī)會(huì)的速度更快，擁有數(shù)據(jù)和生成數(shù)據(jù)存在準(zhǔn)入門檻，這可能會(huì)導(dǎo)致每個(gè)行業(yè)中只有少數(shù)幾家公司占據(jù)主導(dǎo)地位。

Such grandios prediction kindled hope?and?as well as anxiety. Many fret that AI will destroy jobs faster than creat it. Barries of?to entry to have?from owing and generating data may lead to a handful of dominant companies in every industries.

不被熟知但同樣重要的是AI將會(huì)如何改變工作環(huán)境。運(yùn)用AI，主管可以很好的職工實(shí)施控制。亞馬遜公司獲得了一項(xiàng)手環(huán)的專利，追蹤倉(cāng)庫(kù)工人手的移動(dòng)，并用震動(dòng)來(lái)促使他們更加高效。軟件公司W(wǎng)orkday設(shè)置了60種因素來(lái)預(yù)測(cè)哪一個(gè)員工將會(huì)離開。Humanyze售出可以在辦公室周圍追蹤員工并揭示他們?nèi)绾魏屯鹿ぷ鞯闹悄苌矸莼照隆?/i>

人們不太熟悉、但同樣重要的一個(gè)問(wèn)題是AI將如何改變工作場(chǎng)所。利用AI，管理者可以非常嚴(yán)密地控制他們的員工。亞馬遜獲得專利的腕帶可以追蹤倉(cāng)庫(kù)工人手部動(dòng)作，并通過(guò)震動(dòng)敦促他們提高效率。軟件公司W(wǎng)orkdays研究了大約60個(gè)因素來(lái)預(yù)測(cè)哪個(gè)員工會(huì)離職。創(chuàng)業(yè)公司Humanyze銷售的智能工牌可以追蹤員工在辦公室的動(dòng)向，顯示他們與同事的互動(dòng)情況。

Not?Less familiar, but as important, is how AI will transform workplace. Using AI, managers can take a extrordinary control of their employees. Amazon has patented a wristband that can tracks hand's movements of workers in warehouse and uses vibrations to nugde them into being more efficient. Workdays, a software firm, had crunched 60 factors to forecas which worker will leave. Humanyze, a startup, sells smart ID badges to track employees arround the office and reveal how well they work with?interact with colleges colleagues.

在工作室實(shí)施監(jiān)控并不是什么新鮮事。公司的職工很早就要進(jìn)出打卡了；老板可以在他們的電腦上看到是什么讓員工們無(wú)所事事。但是AI使監(jiān)控更加廣泛，因?yàn)槊恳蛔止?jié)的數(shù)據(jù)都可能有潛在的價(jià)值。幾乎沒(méi)有法律說(shuō)明這些數(shù)據(jù)應(yīng)該在工作中如何被收集?？赡芎芏鄦T工在他們簽訂職工合同的時(shí)候不小心地同意了監(jiān)控，這回造成什么問(wèn)題呢？

工作中的監(jiān)控并不是什么新鮮事。工廠工人一直都在打卡上下班；老板們也已經(jīng)能看到無(wú)所事事的員工在電腦上干什么。但是又了AI以后，就值得部屬無(wú)處不在的監(jiān)控了，因?yàn)槊恳粋€(gè)數(shù)據(jù)都有潛在的價(jià)值。很少有法律規(guī)定該如何收集工作數(shù)據(jù)，而許多員工在簽訂雇傭合同時(shí)都會(huì)毫無(wú)戒心地統(tǒng)一接受監(jiān)控，這一切會(huì)導(dǎo)致什么結(jié)果？

Surveillance at work is nothing new. Factory workers have long clocked in and out; bosses can already see what idle workers do on their computers. But AI makes ubiquitous serveillence worthwhile, because every bits of data may have potential worth valuable. Few laws govern how can these data be collected at work, May many employees unguardedly consent to surveillance when signing their employment contract, Where does all this lead？?

信任和監(jiān)視

Trust and surveillance

首先是好處，AI提高了生產(chǎn)力。Humanyze 公司從工牌中整合的關(guān)于員工日歷和郵箱的數(shù)據(jù)反映了公司的布局是否有利于團(tuán)隊(duì)合作。Slack，一個(gè)工作信息傳遞的APP，幫助主管評(píng)估員工完成任務(wù)的速度，公司不僅可以檢測(cè)到員工什么時(shí)候在打瞌睡，而且什么時(shí)候行為不檢。他們開始用AI去檢測(cè)不正常的報(bào)銷，在夜晚的奇數(shù)標(biāo)記收款比人來(lái)做更有效率。

先從好處說(shuō)起。AI應(yīng)該能提高生產(chǎn)率。Humanyze公司將其智能工牌中的數(shù)據(jù)與員工的日程表和電子郵箱信息結(jié)合分析，可以獲得一些發(fā)現(xiàn)，比如辦公室的布局是否有利于團(tuán)隊(duì)合作。職場(chǎng)即時(shí)通訊應(yīng)用Slack幫助管理者評(píng)估員工完成工作的速度。公司不僅能發(fā)現(xiàn)員工打瞌睡，還能看到他們何時(shí)行為不當(dāng)。公司還開始用AI來(lái)篩查費(fèi)用報(bào)銷中的異常，把晚間異常時(shí)段的發(fā)票標(biāo)注出來(lái)，比那些會(huì)算賬的活人效率更高。

Begin with the benefifs, AI ought to improve productivity. Humanyze merges data from smart budges with workers' calendars and e-mails to work out, say, whether office layouts favour teamwork. Slack, a office message app, helps managers assess how quickly the workers accomplish tasks. companies will see what time employees are not only dozing off, but also misbehaving. They start to use AI to screen for anomalies in expense claims, flagging receipts from odd hours of the night more efficiently than a carbon-based beancounter can.

員工也能從中獲益。感謝計(jì)算機(jī)可視化的進(jìn)步，AI可以檢查員工是否裝備安全，并且沒(méi)有人在工廠受傷。一些人會(huì)感謝AI給他們?cè)诠ぷ髦械姆答伈⑶壹?lì)自己做的更好。創(chuàng)業(yè)公司Cogito基于AI設(shè)計(jì)了一個(gè)軟件，用來(lái)監(jiān)聽克服電話并且基于代理人是如何寬容的并多快、多有能力的處理投訴，來(lái)評(píng)估同情分。

員工也會(huì)獲益。有賴于計(jì)算機(jī)視覺(jué)的進(jìn)步，AI可以檢查工人是否串號(hào)了安全設(shè)備，車間是夠有人受傷害等。有些人會(huì)很高興獲得更多有關(guān)自己工作的反饋，并樂(lè)于知道怎樣做更好。創(chuàng)業(yè)公司Cogito設(shè)計(jì)了一款A(yù)I增強(qiáng)軟件，可以監(jiān)聽客服電話，根據(jù)客服的體恤度、解決客戶投訴的速度和能力打出“共情分”。

employees will gain too. Thanks to strides in computer vision, AI can check that workers are wearing safety gear and no that one has been harmed on the factory floor. Some will appreciate more feedback on their work and welcome a sense of how to do better. Cogito designed a AI-enhanced software that listens to custom-service calls and assigns a "sympathy score", based on how compasionate agents are and how quickly and how capably they settle complaints.?

設(shè)備可以保證升職加薪屬于那些值得的人。首先是解雇，人們總是有私心，而合理設(shè)計(jì)的算法可以更加公平。軟件可以指出人可能遺漏的部分。創(chuàng)業(yè)公司Textio用AI來(lái)改進(jìn)工作說(shuō)明，發(fā)現(xiàn)女性對(duì)發(fā)展一個(gè)團(tuán)隊(duì)而不是管理一個(gè)團(tuán)隊(duì)更有興趣。算法將會(huì)排除那些管理者可能有意或無(wú)意看中的性別、種族因素在工資中造成的差異，甚至是性騷擾和種族歧視。

機(jī)器有助于確保加薪和晉升的機(jī)會(huì)留給那些應(yīng)得的人，這從招聘就開始了。人往往有偏見，而且如果設(shè)計(jì)得當(dāng)，算法能比人更公正。軟件可以標(biāo)識(shí)出人可能注意不到的模式。Textio是一家利用AI改進(jìn)崗位描述的創(chuàng)業(yè)公司，他發(fā)現(xiàn)女性更有可能應(yīng)征那種談及“發(fā)展”團(tuán)隊(duì)而非“管理”團(tuán)隊(duì)的工作。算法會(huì)發(fā)現(xiàn)不同性別和種族間的薪酬差異，以及人類管理者有意無(wú)意忽略的性騷擾和種族歧視。

Machines can ensure that pay rises and promotions go to those who deserve them. That begin with hiring. People always have biases but algorithms, if designed properlycorrectly, can be more impartial. Software can flag patterns people miss. Textio, a startup thay use AI to improve job descriptions, revealed that women are likelier to respond to jobs that mention developing a team rather than managing a team. Algorithm will pick up differencesin pay between genders and races,as well as sexual harassment and racism that human managers conciously or unconsciously overlook.

AI帶來(lái)好處的同時(shí)也帶來(lái)了潛在的缺陷。算法可能會(huì)包含程序員的偏見，也可能會(huì)有不注意的原因。上下班的路程可能能夠預(yù)測(cè)一個(gè)員工是否會(huì)辭職，但是這可能會(huì)非故意的傷害到貧窮的求職者。年齡大的員工可能比年輕的員工工作更慢，如果AI只關(guān)注于生產(chǎn)力會(huì)使他們丟掉工作。

不過(guò)人工智能帶來(lái)好處的同時(shí)也會(huì)有許多潛在的缺陷。算法也許難免會(huì)受程序員偏見的影響。他們也會(huì)產(chǎn)生意想不到的后果。AI或許可以通過(guò)上下班路程的長(zhǎng)短預(yù)測(cè)員工是否會(huì)辭職，但關(guān)注這一點(diǎn)可能會(huì)無(wú)意中傷害到經(jīng)濟(jì)狀況較差的求職者。年長(zhǎng)的員工可能比年輕員工手腳慢，如果AI只關(guān)注生產(chǎn)力，那他們可能就會(huì)丟掉飯碗。

AI 's benefits come with potential drawbacks. Algorithm will not be free of biases of their programmers. They will have unintended consequences. AI may can use a?The length of a commute may predict whether an emplyee will leave, but the focus will inadvertently harm poorer applicants. Older employee will work more slowly than younger ones, it will?and could risk losing their promotions if all AI concerns is productivity.

監(jiān)視可能會(huì)造成奧威爾效應(yīng)，人們開始質(zhì)疑臉書和其他科技代理公司了解多少他們的生活。公司開始檢測(cè)員工休息多長(zhǎng)時(shí)間。軟件公司Veriato，追蹤和檢測(cè)員工按下的每一次鍵盤為了無(wú)測(cè)量他們對(duì)公司有多賣力。公司不僅可以用AI去篩選員工的職業(yè)交流，而且可以檢測(cè)出他們的社交側(cè)寫。從Slack的名字就可以看出，代表了所有談話和知識(shí)的可調(diào)查。

而且監(jiān)控可能會(huì)讓人身處奧威爾筆下的世界。這可是一個(gè)敏感的問(wèn)題——人們已經(jīng)開始質(zhì)疑Facebook和其他科技巨頭究竟掌握了多少有關(guān)他們的私生活內(nèi)容。公司開始監(jiān)控員工花多長(zhǎng)時(shí)間在工間休息上。軟件公司Veriato的產(chǎn)品甚至追蹤和記錄員工在電腦上的一舉一動(dòng)，以此來(lái)衡量他們對(duì)雇主是否盡心盡力。如今企業(yè)不僅可以利用AI篩查員工的工作通信，還能詳查他們的社交媒體信息。Slack這個(gè)名字就很能說(shuō)明問(wèn)題——它是一個(gè)縮寫，代表“所有的對(duì)話和信息的可搜索日志”。

Surveillance will feel Orwellian——a sensitive matter that people have begun to question how much Facebook and other tech giants learn about their privacy lives. Firms start to screen?monitor how much time spend on workers breaks. Veriato, a software firm, goes so far as to track and log every keystroke employees make on their computer to gauge how committed the workers are to the company. Companies can use AI to?shift through not just?workers office?professional communications, but their social-media profiles. The Slack's name is the clue, which stands for “Searchable log of all communcation and knowledge”.

跟蹤追蹤器

Tracking the tracker

一些人可以比其他門更好的阻止管理者做的太過(guò)分。如果技能被需要。你比容易替代的人更能夠抵抗這些監(jiān)視。低工資領(lǐng)域的按小時(shí)計(jì)費(fèi)的工作，比如零售，可能更加脆弱。這可能會(huì)推動(dòng)工會(huì)的死灰復(fù)燃，去代表員工的利益并設(shè)置一些規(guī)定。即使這樣，一些職位就在被機(jī)器人代替或被像一個(gè)機(jī)器人一樣對(duì)待中選擇。

一些人會(huì)比其他人更有底氣讓雇主不要做得太過(guò)頭。如果市場(chǎng)需要你的技能，那么你就比那些可被輕易取代的人更可能抵擋住這種趨勢(shì)。在零售業(yè)等低薪行業(yè)領(lǐng)取小時(shí)工資的員工處境尤其不妙。這可能會(huì)促使力圖代表員工利益、尋求指定規(guī)范的工會(huì)復(fù)興。但即便真的如此，在某些崗位上將只有這樣的選擇：要么被機(jī)器人取代，要么被當(dāng)機(jī)器人對(duì)待。

Some are placed better than others to stop managers going too far. If your skills are in demand, the surveiliance can?You are more likely to resist than if you are easy to replace. paid-in-hours paid-by-the-hour workers in low-wage industries such as retailing may be more?especially vulnerable. It may fuel resurgence of labour unions seeking to represent worker's interests and to set norms. Although?Even then, the choice of some jobs is between being replaced by a robot or being treated like one.

當(dāng)管理者和被雇傭者權(quán)衡AI在工作上的好處和缺點(diǎn)，三條原則應(yīng)該去指導(dǎo)AI的蔓延。首先，數(shù)據(jù)應(yīng)該盡可能的匿名。比如說(shuō)微軟公司，擁有一個(gè)產(chǎn)品，顯示個(gè)人在辦公室如何管理時(shí)間，但是只將聚合形式的信息給管理者。第二，AI的使用應(yīng)該變得透明。員工更應(yīng)該被告知工作場(chǎng)所在使用干什么技術(shù)，并且什么數(shù)據(jù)會(huì)被收集。作為慣例，用于公司招聘，解雇和升職的算法應(yīng)該測(cè)試偏見和不注意的因素。最后，國(guó)家應(yīng)該允許個(gè)體得到他們的數(shù)據(jù)，無(wú)論他們是想要找到解雇原因的前公司職員，還會(huì)希望向可能的雇傭者說(shuō)明自身能力的找工作的人。

監(jiān)管機(jī)構(gòu)和雇主在權(quán)衡職場(chǎng)應(yīng)用AI的利弊時(shí)，應(yīng)采用三大原則來(lái)指導(dǎo)這項(xiàng)技術(shù)的普及。首先，在可能的情況下，數(shù)據(jù)應(yīng)該匿名。例如微軟有個(gè)產(chǎn)品，可向個(gè)人展示他們?nèi)绾喂芾碜约旱墓ぷ鲿r(shí)間，但只向經(jīng)理們提供整合后的信息。其次，AI的使用應(yīng)該是透明的。公司應(yīng)告知員工正在使用哪些技術(shù)收集哪些數(shù)據(jù)。公司用于雇傭、解雇和晉升決策的算法應(yīng)接受測(cè)試，看是否存在偏見或意外后果，而且這應(yīng)成為常規(guī)做法。最后，各國(guó)應(yīng)該允許個(gè)人索要自己的數(shù)據(jù)，不管他們是想就被解雇提出異議的前員工，還是希望向未來(lái)雇主展示自己能力的求職者。

As regulators and employers weight the pros and cons of AI at work, three principles ought to guide the spread. First, data should be annoymised where possible. Such as Microsoft, for exemple, has a product that show individuals how they manager their time in the office, but offer managers information in aggerated form. Second, the use of AI should be transparent. Companies should tell workers what techologies are being used in their workplaces and which data are being gathered. As normal routineAs a matter of routine, Algorithm used by firm to hire, fire and promote should be tested for bias and unintented consequences. Finally, countries shouls let individuals request their own data, whether they are ex-workers wanting to contest a dismissal or jobseekers hoping to demonstrate their ability to prospective employers.

AI進(jìn)入職場(chǎng)需要權(quán)衡隱私和表現(xiàn)。一個(gè)更加公平和高產(chǎn)的工作推動(dòng)力是一個(gè)值得擁有的獎(jiǎng)勵(lì)，但要它不是束縛和反人道主義的。保持其中的平衡需要思考，需要雇傭者和被雇傭者共同的適應(yīng)，和人道主義的需求。

AI大舉進(jìn)軍工作場(chǎng)所，人們需要在隱私和績(jī)效之間權(quán)衡取舍。一個(gè)更公平、生產(chǎn)率更高的工作場(chǎng)所值得擁有，但如果它束縛了員工，抹殺了人性，就得不償失了。要在兩者間取得平衡需要深思熟慮，需要雇主和雇員都有調(diào)適的意愿，還需要大量的人性關(guān)懷。

AI enter work should?The march of AI into workplace calls for trade-offs?between privacy and performance. A fairer and productive workforce is a prize worth having, but not if it shackles and dehumanises employees. Stricking a balance need thought, a willingness for both employers and emplyees to adapt, and a stronge dose of humanity.

參考翻譯：APP????經(jīng)濟(jì)學(xué)人·商論

原文：經(jīng)濟(jì)學(xué)人20180329 ? ?AI-spy: The workplace of the future