三大突破讓人工智能終成現實

A few months ago I made the trek to the sylvan campus of the IBM research labs in Yorktown Heights, New York, to catch an early glimpse of the fast-arriving, long-overdue future of artificial intelligence. This was the home of Watson, the electronic genius that conquered Jeopardy! in 2011. The original Watson is still here—it's about the size of a bedroom, with 10 upright, refrigerator-shaped machines forming the four walls. The tiny interior cavity gives technicians access to the jumble of wires and cables on the machines' backs. It is surprisingly warm inside, as if the cluster were alive.

數月前，我長途跋涉來到位於紐約州約克城高地的IBM研究實驗室的林間園區，爲的就是能早早一窺那近在眼前卻讓人期待許久的人工智能的未來。這兒是超級電腦“沃森”（Watson）的研發地，而沃森在2011年就在“危險邊緣”（Jeopardy！）節目的比賽裏拔得頭籌。最初的沃森電腦仍留於此處——它是一個體積約與一個臥室相當，由10臺直立的冷櫃式機器圍成四面牆的計算機系統。技術人員可以通過系統內部的小細孔把各種線纜接到機器背部。系統內部溫度高得出奇，彷彿這個計算機集羣是活生生的一般。

Today's Watson is very different. It no longer exists solely within a wall of cabinets but is spread across a cloud of open-standard servers that run several hundred “instances” of the AI at once. Like all things cloudy, Watson is served to simultaneous customers anywhere in the world, who can access it using their phones, their desktops, or their own data servers. This kind of AI can be scaled up or down on demand. Because AI improves as people use it, Watson is always getting smarter; anything it learns in one instance can be immediately transferred to the others. And instead of one single program, it's an aggregation of diverse software engines—its logic-deduction engine and its language-parsing engine might operate on different code, on different chips, in different locations—all cleverly integrated into a unified stream of intelligence.

如今的沃森系統與之前相比有了顯著差異。它不再僅僅存在於一排機櫃之中，而是通過大量對用戶免費開放的服務器傳播，這些服務器能夠即時運行上百種人工智能的“情況”。同所有云端化的事物一樣，沃森系統爲世界各地同時使用的客戶服務，他們能夠用手機、臺式機以及他們自己的數據服務器連上該系統。這類人工智能可以根據需求按比例增加或減少。鑑於人工智能會隨人們的使用而逐步改進，沃森將始終變得愈發聰明；它在任何一次情況中所獲悉的改進點都會立即傳送至其他情況中。並且，它也不是一個單一的程序，而是各種軟件引擎的集合——其邏輯演繹引擎和語言解析引擎可以在不同的代碼、芯片以及位置上運行——所有這些智慧的因素都彙集成了一個統一的智能流。

Consumers can tap into that always-on intelligence directly, but also through third-party apps that harness the power of this AI cloud. Like many parents of a bright mind, IBM would like Watson to pursue a medical career, so it should come as no surprise that one of the apps under development is a medical-diagnosis tool. Most of the previous attempts to make a diagnostic AI have been pathetic failures, but Watson really works. When, in plain English, I give it the symptoms of a disease I once contracted in India, it gives me a list of hunches, ranked from most to least probable. The most likely cause, it declares, is Giardia—the correct answer. This expertise isn't yet available to patients directly; IBM provides access to Watson's intelligence to partners, helping them develop user-friendly interfaces for subscribing doctors and hospitals. “I believe something like Watson will soon be the world's best diagnostician—whether machine or human,” says Alan Greene, chief medical officer of Scanadu, a startup that is building a diagnostic device inspired by the Star Trek medical tricorder and powered by a cloud AI. “At the rate AI technology is improving, a kid born today will rarely need to see a doctor to get a diagnosis by the time they are an adult.”

用戶可以直接接入這一永久連接（always-on）的智能系統，也可以通過使用這一人工智能雲服務的第三方應用程序接入。正如許多高瞻遠矚的父母一樣，IBM想讓沃森電腦從事醫學工作，因此他們正在開發一款醫療診斷工具的應用程序，這倒也不足爲奇。之前，診療方面的人工智能嘗試大多以慘敗告終，但沃森卻卓有成效。簡單地說，當我輸入我曾經在印度感染上的某種疾病症狀時，它會給我一個疑似病症的清單，上面一一列明瞭可能性從高到低的疾病。它認爲我最可能感染了賈第鞭毛蟲病（Giardia）——說的一點兒也沒錯。這一技術尚未直接對患者開放；IBM將沃森電腦的智能提供給合作伙伴接入使用，以幫助他們開發出用戶友好界面爲預約醫生及醫院方面服務。“我相信類似沃森這種——無論它是機器還是人——都將很快成爲世界上最好的診療醫生”，創業公司Scanadu的首席醫療官艾倫·格林（Alan Greene）說道，該公司受到電影《星際迷航》中醫用三錄儀的啓發，正在利用雲人工智能技術製造一種診療設備。“從人工智能技術改進的速率來看，現在出生的孩子長大後，很可能不太需要通過看醫生來得知診療情況了。”

As AIs develop, we might have to engineer ways to prevent consciousness in them—our most premium AI services will be advertised as consciousness-free.

隨着人工智能發展，我們可能要設計出一些阻止它們擁有意識的方式——我們所宣稱的最優質的人工智能服務將是無意識服務。

Medicine is only the beginning. All the major cloud companies, plus dozens of startups, are in a mad rush to launch a Watson-like cognitive service. According to quantitative analysis firm Quid, AI has attracted more than $17 billion in investments since 2009. Last year alone more than $2 billion was invested in 322 companies with AI-like technology. Facebook and Google have recruited researchers to join their in-house AI research teams. Yahoo, Intel, Dropbox, LinkedIn, Pinterest, and Twitter have all purchased AI companies since last year. Private investment in the AI sector has been expanding 62 percent a year on average for the past four years, a rate that is expected to continue.

醫學僅僅只是一個開始。所有主流的雲計算公司，加上數十家創業公司都在爭先恐後地開展類似沃森電腦的認知服務。根據量化分析公司Quid的數據，自2009年以來，人工智能已經吸引了超過170億美元的投資。僅去年一年，就有322家擁有類似人工智能技術的公司獲得了超過20億美元的投資。Facebook和谷歌也爲其公司內部的人工智能研究小組招聘了研究員。自去年以來，雅虎、英特爾、Dropbox、LinkedIn、Pinterest以及推特也都收購了人工智能公司。過去四年間，人工智能領域的民間投資以平均每年62%的增長速率增加，這一速率預計還會持續下去。

Amid all this activity, a picture of our AI future is coming into view, and it is not the HAL 9000—a discrete machine animated by a charismatic (yet potentially homicidal) humanlike consciousness—or a Singularitan rapture of superintelligence. The AI on the horizon looks more like Amazon Web Services—cheap, reliable, industrial-grade digital smartness running behind everything, and almost invisible except when it blinks off. This common utility will serve you as much IQ as you want but no more than you need. Like all utilities, AI will be supremely boring, even as it transforms the Internet, the global economy, and civilization. It will enliven inert objects, much as electricity did more than a century ago. Everything that we formerly electrified we will now cognitize. This new utilitarian AI will also augment us individually as people (deepening our memory, speeding our recognition) and collectively as a species. There is almost nothing we can think of that cannot be made new, different, or interesting by infusing it with some extra IQ. In fact, the business plans of the next 10,000 startups are easy to forecast: Take X and add AI. This is a big deal, and now it's here.

縱觀所有這些活動，人工智能的未來正進入我們的視野之中，它既非如那種哈爾9000（HAL 9000）（譯者注：小說及電影《2001：太空漫遊》中的超級電腦）——一臺擁有超凡（但有潛在嗜殺傾向）的類人意識並依靠此運行的獨立機器那般——也非讓奇點論者心醉神迷的超級智能。即將到來的人工智能頗似亞馬遜的網絡服務——廉價、可靠、工業級的數字智慧在一切事物的背後運行，偶爾在你的眼前閃爍幾下，其他時候近乎無形。這一通用設施將提供你所需要的人工智能而不超出你的需要。和所有設施一樣，即使人工智能改變了互聯網、全球經濟以及文明，它也將變得令人厭倦。正如一個多世紀以前電力所做的那樣，它會讓無生命的物體活躍起來。之前我們電氣化的所有東西，現在我們都將使之認知化。而實用化的新型人工智能也會增強人類個體（加深我們的記憶、加速我們的認知）以及人類羣體的生活。通過加入一些額外的智能因素，我們想不到有什麼東西不能變得新奇、不同且有趣。實際上，我們能輕易地預測到接下來的一萬家創業公司的商業計劃：“做某項事業，並加入人工智能”。茲事體大，近在眼前。

Around 2002 I attended a small party for Google—before its IPO, when it only focused on search. I struck up a conversation with Larry Page, Google's brilliant cofounder, who became the company's CEO in 2011. “Larry, I still don't get it. There are so many search companies. Web search, for free? Where does that get you?” My unimaginative blindness is solid evidence that predicting is hard, especially about the future, but in my defense this was before Google had ramped up its ad-auction scheme to generate real income, long before YouTube or any other major acquisitions. I was not the only avid user of its search site who thought it would not last long. But Page's reply has always stuck with me: “Oh, we're really making an AI.”

大約在2002年時，我參加了谷歌的一個小型聚會——彼時谷歌尚未IPO，還在一心一意地做網絡搜索。我與谷歌傑出的聯合創始人、2011年成爲谷歌CEO的拉里·佩奇（Larry Page）隨意攀談起來。“拉里，我還是搞不懂，現在有這麼多搜索公司，你們爲什麼要做免費的網絡搜索？你是怎麼想到這個主意的？”我那缺乏想象力的無知着實證明了我們很難去做預測，尤其是對於未來的預測。但我要辯解的是，在谷歌增強其廣告拍賣方案並使之形成實際收益，以及進行對YouTube的併購或其他重要併購之前，預測未來是很難的。我並不是唯一一個一邊狂熱地用着谷歌的搜索引擎一邊認爲它撐不了多久的用戶。但佩奇的回答讓我一直難以忘懷：“哦，我們實際上是在做人工智能。”

I've thought a lot about that conversation over the past few years as Google has bought 14 AI and robotics companies. At first glance, you might think that Google is beefing up its AI portfolio to improve its search capabilities, since search contributes 80 percent of its revenue. But I think that's backward. Rather than use AI to make its search better, Google is using search to make its AI better. Every time you type a query, click on a search-generated link, or create a link on the web, you are training the Google AI. When you type “Easter Bunny” into the image search bar and then click on the most Easter Bunny-looking image, you are teaching the AI what an Easter bunny looks like. Each of the 12.1 billion queries that Google's 1.2 billion searchers conduct each day tutor the deep-learning AI over and over again. With another 10 years of steady improvements to its AI algorithms, plus a thousand-fold more data and 100 times more computing resources, Google will have an unrivaled AI. My prediction: By 2024, Google's main product will not be search but AI.

過去數年間，關於那次談話我想了很多，谷歌也收購了14家人工智能以及機器人方面的公司。鑑於搜索業務爲谷歌貢獻了80%的收入，因此乍一看去，你可能會覺得谷歌正在擴充其人工智能方面的投資組合以改善搜索能力。但是我認爲正好相反。谷歌正在用搜索技術來改善人工智能，而非使用人工智能來改進搜索技術。每當你輸入一個查詢詞，點擊搜索引擎生成的鏈接，或者在網頁上創造一個鏈接，你都是在訓練谷歌的人工智能技術。當你在圖片搜索欄中輸入“復活節兔子”（Easter Bunny）並點擊看起來最像復活節兔子的那張圖片時，你都是在告訴人工智能，復活節兔子是長成什麼樣的。谷歌每天擁有12億搜索用戶，產生1210億搜索關鍵詞，每一個關鍵詞都是在一次又一次地輔導人工智能進行深度學習。如果再對人工智能的算法進行爲之10年的穩固改進，加之一千倍以上的數據以及一百倍以上的計算資源，谷歌將會開發出一款無與倫比的人工智能產品。我的預言是：到2024年，谷歌的主營產品將不再是搜索引擎，而是人工智能產品。

This is the point where it is entirely appropriate to be skeptical. For almost 60 years, AI researchers have predicted that AI is right around the corner, yet until a few years ago it seemed as stuck in the future as ever. There was even a term coined to describe this era of meager results and even more meager research funding: the AI winter. Has anything really changed?

這個觀點自然也會招來懷疑的聲音。近60年來，人工智能的研究者都預測說人工智能時代即將到來，但是直到幾年前，人工智能好像還是遙不可及。人們甚至發明了一個詞來描述這個研究結果匱乏、研究基金更加匱乏的時代：人工智能之冬。那麼事情真的有變化嗎？

Yes. Three recent breakthroughs have unleashed the long-awaited arrival of artificial intelligence:

是的。近期的三大突破讓人們期待已久的人工智能近在眼前：

1. Cheap parallel computation

1. 成本低廉的並行計算

Thinking is an inherently parallel process, billions of neurons firing simultaneously to create synchronous waves of cortical computation. To build a neural network—the primary architecture of AI software—also requires many different processes to take place simultaneously. Each node of a neural network loosely imitates a neuron in the brain—mutually interacting with its neighbors to make sense of the signals it receives. To recognize a spoken word, a program must be able to hear all the phonemes in relation to one another; to identify an image, it needs to see every pixel in the context of the pixels around it—both deeply parallel tasks. But until recently, the typical computer processor could only ping one thing at a time.

思考是一種人類固有的並行過程，數以億計的神經元同時放電以創造出大腦皮層用於計算的同步腦電波。搭建一個神經網絡——即人工智能軟件的主要結構——也需要許多不同的進程同時運行。神經網絡的每一個節點都大致模擬了大腦中的一個神經元——其與相鄰的節點互相作用，以明確所接收的信號。一項程序要理解某個口語單詞，就必須能夠聽清（不同音節）彼此之間的所有音素；要識別出某幅圖片，就需要看到其周圍像素環境內的所有像素——二者都是深層次的並行任務。但直到最近，標準的計算機處理器也僅僅能一次處理一項任務。

That began to change more than a decade ago, when a new kind of chip, called a graphics processing unit, or GPU, was devised for the intensely visual—and parallel—demands of videogames, in which millions of pixels had to be recalculated many times a second. That required a specialized parallel computing chip, which was added as a supplement to the PC motherboard. The parallel graphical chips worked, and gaming soared. By 2005, GPUs were being produced in such quantities that they became much cheaper. In 2009, Andrew Ng and a team at Stanford realized that GPU chips could run neural networks in parallel.

事情在十多年前就已經開始發生變化，彼時出現了一種被稱爲圖形處理單元（graphics processing unit -GPU）的新型芯片，它能夠滿足可視遊戲中高密度的視覺以及並行需求，在這一過程中，每秒鐘都有上百萬像素被多次重新計算。這一過程需要一種專門的並行計算芯片，該芯片添加至電腦主板上，作爲對其的補充。並行圖形芯片作用明顯，遊戲可玩性也大幅上升。到2005年，GPU芯片產量頗高，其價格便降了下來。2009年，吳恩達（Andrew Ng）（譯者注：華裔計算機科學家）以及斯坦福大學的一個研究小組意識到，GPU芯片可以並行運行神經網絡。

That discovery unlocked new possibilities for neural networks, which can include hundreds of millions of connections between their nodes. Traditional processors required several weeks to calculate all the cascading possibilities in a 100 million-parameter neural net. Ng found that a cluster of GPUs could accomplish the same thing in a day. Today neural nets running on GPUs are routinely used by cloud-enabled companies such as Facebook to identify your friends in photos or, in the case of Netflix, to make reliable recommendations for its more than 50 million subscribers.

這一發現開啓了神經網絡新的可能性，使得神經網絡能容納上億個節點間的連接。傳統的處理器需要數週才能計算出擁有1億節點的神經網的級聯可能性。而吳恩達發現，一個GPU集羣在一天內就可完成同一任務。現在，一些應用雲計算的公司通常都會使用GPU來運行神經網絡，例如，Facebook會籍此技術來識別用戶照片中的好友，Netfilx也會依其來給5000萬訂閱用戶提供靠譜的推薦內容。

2. Big Data

2. 大數據

Every intelligence has to be taught. A human brain, which is genetically primed to categorize things, still needs to see a dozen examples before it can distinguish between cats and dogs. That's even more true for artificial minds. Even the best-programmed computer has to play at least a thousand games of chess before it gets good. Part of the AI breakthrough lies in the incredible avalanche of collected data about our world, which provides the schooling that AIs need. Massive databases, self-tracking, web cookies, online footprints, terabytes of storage, decades of search results, Wikipedia, and the entire digital universe became the teachers making AI smart.

每一種智能都需要被訓練。哪怕是天生能夠給事物分類的人腦，也仍然需要看過十幾個例子後才能夠區分貓和狗。人工思維則更是如此。即使是（國際象棋）程序編的最好的電腦，也得在至少對弈一千局之後纔能有良好表現。人工智能獲得突破的部分原因在於，我們收集到來自全球的海量數據，以給人工智能提供了其所需的訓練。巨型數據庫、自動跟蹤（self-tracking）、網頁cookie、線上足跡、兆兆字節級存儲、數十年的搜索結果、維基百科以及整個數字世界都成了老師，是它們讓人工智能變得更加聰明。

3. Better algorithms

3. 更優的算法

Digital neural nets were invented in the 1950s, but it took decades for computer scientists to learn how to tame the astronomically huge combinatorial relationships between a million—or 100 million—neurons. The key was to organize neural nets into stacked layers. Take the relatively simple task of recognizing that a face is a face. When a group of bits in a neural net are found to trigger a pattern—the image of an eye, for instance—that result is moved up to another level in the neural net for further parsing. The next level might group two eyes together and pass that meaningful chunk onto another level of hierarchical structure that associates it with the pattern of a nose. It can take many millions of these nodes (each one producing a calculation feeding others around it), stacked up to 15 levels high, to recognize a human face. In 2006, Geoff Hinton, then at the University of Toronto, made a key tweak to this method, which he dubbed “deep learning.” He was able to mathematically optimize results from each layer so that the learning accumulated faster as it proceeded up the stack of layers. Deep-learning algorithms accelerated enormously a few years later when they were ported to GPUs. The code of deep learning alone is insufficient to generate complex logical thinking, but it is an essential component of all current AIs, including IBM's Watson, Google's search engine, and Facebook's algorithms.

20世紀50年代，數字神經網絡就被髮明瞭出來，但計算機科學家花費了數十年來研究如何駕馭百萬乃至億級神經元之間那龐大到如天文數字一般的組合關係。這一過程的關鍵是要將神經網絡組織成爲堆疊層（stacked layer）。一個相對來說比較簡單的任務就是人臉識別。當某神經網絡中的一組比特被發現能夠形成某種圖案——例如，一隻眼睛的圖像——這一結果就會被向上轉移至該神經網絡的另一層以做進一步分析。接下來的這一層可能會將兩隻眼睛拼在一起，將這一有意義的數據塊傳遞到層級結構的第三層，該層可以將眼睛和鼻子的圖像結合到一起（來進行分析）。識別一張人臉可能需要數百萬個這種節點（每個節點都會生成一個計算結果以供周圍節點使用），並需要堆疊高達15個層級。2006年，當時就職於多倫多大學的傑夫·辛頓（Geoff Hinton）對這一方法進行了一次關鍵改進，並將其稱之爲“深度學習”。他能夠從數學層面上優化每一層的結果從而使神經網絡在形成堆疊層時加快學習速度。數年後，當深度學習算法被移植到GPU集羣中後，其速度有了顯著提高。僅靠深度學習的代碼並不足以能產生複雜的邏輯思維，但是它是包括IBM的沃森電腦、谷歌搜索引擎以及Facebook算法在內，當下所有人工智能產品的主要組成部分。

This perfect storm of parallel computation, bigger data, and deeper algorithms generated the 60-years-in-the-making overnight success of AI. And this convergence suggests that as long as these technological trends continue—and there's no reason to think they won't—AI will keep improving.

這一由並行計算、大數據和更深層次算法組成的完美風暴使得持續耕耘了60年的人工智能一鳴驚人。而這一聚合也表明，只要這些技術趨勢繼續下去——它們也沒有理由不延續——人工智能將精益求精。

As it does, this cloud-based AI will become an increasingly ingrained part of our everyday life. But it will come at a price. Cloud computing obeys the law of increasing returns, sometimes called the network effect, which holds that the value of a network increases much faster as it grows bigger. The bigger the network, the more attractive it is to new users, which makes it even bigger, and thus more attractive, and so on. A cloud that serves AI will obey the same law. The more people who use an AI, the smarter it gets. The smarter it gets, the more people use it. The more people that use it, the smarter it gets. Once a company enters this virtuous cycle, it tends to grow so big, so fast, that it overwhelms any upstart competitors. As a result, our AI future is likely to be ruled by an oligarchy of two or three large, general-purpose cloud-based commercial intelligences.

隨着這一趨勢的持續，這種基於雲技術的人工智能將愈發成爲我們日常生活中不可分割的一部分。但天上沒有掉餡餅的事。雲計算遵循收益遞增（increasing returns）法則，這一法則有時也被稱爲網絡效應（network effect），即隨着網絡發展壯大，網絡價值也會以更快的速度增加。網絡（規模）越大，對於新用戶的吸引力越強，這又讓網絡變得更大，又進一步增強了吸引力，如此往復。爲人工智能服務的雲技術也遵循這一法則。越多人使用人工智能產品，它就會變得越聰明；它變得越聰明，就有越多人來使用它；然後它變得更聰明，進一步就有更多人使用它。一旦有公司邁進了這個良性循環中，其規模會變大、發展會加快，以至於沒有任何新興對手能望其項背。因此，人工智能的未來將有兩到三家寡頭公司統治，它們會開發出大規模基於雲技術的多用途商業智能產品。

In 1997, Watson's precursor, IBM's Deep Blue, beat the reigning chess grand master Garry Kasparov in a famous man-versus-machine match. After machines repeated their victories in a few more matches, humans largely lost interest in such contests. You might think that was the end of the story (if not the end of human history), but Kasparov realized that he could have performed better against Deep Blue if he'd had the same instant access to a massive database of all previous chess moves that Deep Blue had. If this database tool was fair for an AI, why not for a human? To pursue this idea, Kasparov pioneered the concept of man-plus-machine matches, in which AI augments human chess players rather than competes against them.

1997年，沃森電腦的前輩、IBM公司的深藍電腦在一場著名的人機大賽中擊敗了當時的國際象棋大師加里·卡斯帕羅夫（Garry Kasparov）。在電腦又贏了幾場比賽之後，人們基本上失去了對這類比賽的興趣。你可能會認爲故事到此就結束了，但卡斯帕羅夫意識到，如果他也能像深藍一樣立即訪問包括以前所有棋局棋路變化在內的巨型數據庫的話，他在對弈中能表現得更好。如果這一數據庫工具對於人工智能設備來說是公平的話，爲什麼人類不能使用它呢？爲了探究這一想法，卡斯帕羅夫率先提出了“人加機器”（man-plus-machine）比賽的概念，即用人工智能增強國際象棋選手水平，而非讓人與機器之間對抗。

Now called freestyle chess matches, these are like mixed martial arts fights, where players use whatever combat techniques they want. You can play as your unassisted human self, or you can act as the hand for your supersmart chess computer, merely moving its board pieces, or you can play as a “centaur,” which is the human/AI cyborg that Kasparov advocated. A centaur player will listen to the moves whispered by the AI but will occasionally override them—much the way we use GPS navigation in our cars. In the championship Freestyle Battle in 2014, open to all modes of players, pure chess AI engines won 42 games, but centaurs won 53 games. Today the best chess player alive is a centaur: Intagrand, a team of humans and several different chess programs.

這種比賽如今被稱爲自由式國際象棋比賽，它有點兒像混合武術對抗賽，選手們可以使用任何他們想要用的作戰技巧。你可以單打獨鬥；也可以接受你那裝有超級聰明的國際象棋軟件的電腦給出的幫助，你要做的僅僅是按照它的建議來移動棋子；或者你可以當一個卡斯帕羅夫所提倡的那種“半人半機”的選手。半人半機選手會聽取人工智能設備在其耳邊提出的棋路建議，但是也間或不會採用這些建議——頗似我們開車時候用的GPS導航一般。在接受任何模式選手參賽的2014年自由式國際象棋對抗錦標賽上，純人工智能的國際象棋引擎贏得了42場比賽，而半人半機選手則贏得了53場。當今世上最優秀的國際象棋選手就是半人半機選手Intagrand，它是一個由多人以及數個不同國際象棋程序所組成的小組。

But here's the even more surprising part: The advent of AI didn't diminish the performance of purely human chess players. Quite the opposite. Cheap, supersmart chess programs inspired more people than ever to play chess, at more tournaments than ever, and the players got better than ever. There are more than twice as many grand masters now as there were when Deep Blue first beat Kasparov. The top-ranked human chess player today, Magnus Carlsen, trained with AIs and has been deemed the most computer-like of all human chess players. He also has the highest human grand master rating of all time.

但最令人驚訝的是：人工智能的出現並未讓純人類的國際象棋棋手的水平下降。恰恰相反，廉價、超級智能的國際象棋軟件吸引了更多人來下國際象棋，比賽比以前增多了，棋手的水平也比以前上升了。現在的國際象棋大師（譯者注：國際象棋界的一種等級）人數是深藍戰勝卡斯帕羅夫那時候的兩倍多。現在的排名第一的人類國際象棋棋手馬格努斯·卡爾森（Magnus Carlsen）就曾接受人工智能的訓練，他被認爲是所有人類國際象棋棋手中最接近電腦的棋手，同時也是有史以來積分最高的人類國際象棋大師。

If AI can help humans become better chess players, it stands to reason that it can help us become better pilots, better doctors, better judges, better teachers. Most of the commercial work completed by AI will be done by special-purpose, narrowly focused software brains that can, for example, translate any language into any other language, but do little else. Drive a car, but not converse. Or recall every pixel of every video on YouTube but not anticipate your work routines. In the next 10 years, 99 percent of the artificial intelligence that you will interact with, directly or indirectly, will be nerdily autistic, supersmart specialists.

如果人工智能能幫助人類成爲更優秀的國際象棋棋手，那麼它也能幫助我們成爲更爲優秀的飛行員、醫生、法官以及教師。大多數由人工智能完成的商業工作都將是有專門目的的工作，嚴格限制在智能軟件能做到的工作之內，比如，（人工智能產品）把某種語言翻譯成另一種語言，但卻不能翻譯成第三種語言。再比如，它們可以開車，但卻不能與人交談。或者是能回憶起YouTube上每個視頻的每個像素，卻無法預測你的日常工作。在未來十年，你與之直接或者間接互動的人工智能產品，有99%都將是高度專一、極爲聰明的“專家”。

In fact, this won't really be intelligence, at least not as we've come to think of it. Indeed, intelligence may be a liability—especially if by “intelligence” we mean our peculiar self-awareness, all our frantic loops of introspection and messy currents of self-consciousness. We want our self-driving car to be inhumanly focused on the road, not obsessing over an argument it had with the garage. The synthetic Dr. Watson at our hospital should be maniacal in its work, never wondering whether it should have majored in English instead. As AIs develop, we might have to engineer ways to prevent consciousness in them—and our most premium AI services will likely be advertised as consciousness-free.

實際上，這並非真正的智能，至少不是我們細細想來的那種智能。的確，智能可能是一種傾向——尤其是如果我們眼中的智能意味着我們那特有的自我意識、一切我們所有的那種狂亂的自省循環以及凌亂的自我意識流的話。我們希望無人駕駛汽車能一心一意在路上行駛，而不是糾結於之前和車庫的爭吵。醫院中的綜合醫生“沃森”能專心工作，不要去想自己是不是應該專攻英語。隨着人工智能的發展，我們可能要設計出一些阻止它們擁有意識的方式——我們所宣稱的最優質的人工智能服務將是無意識服務。

What we want instead of intelligence is artificial smartness. Unlike general intelligence, smartness is focused, measurable, specific. It also can think in ways completely different from human cognition. A cute example of this nonhuman thinking is a cool stunt that was performed at the South by Southwest festival in Austin, Texas, in March of this year. IBM researchers overlaid Watson with a culinary database comprising online recipes, USDA nutritional facts, and flavor research on what makes compounds taste pleasant. From this pile of data, Watson dreamed up novel dishes based on flavor profiles and patterns from existing dishes, and willing human chefs cooked them. One crowd favorite generated from Watson's mind was a tasty version of fish and chips using ceviche and fried plantains. For lunch at the IBM labs in Yorktown Heights I slurped down that one and another tasty Watson invention: Swiss/Thai asparagus quiche. Not bad! It's unlikely that either one would ever have occurred to humans.

我們想要的不是智能，而是人工智慧。與一般的智能不同，智慧（產品）具有專心、可衡量、種類特定的特點。它也能夠以完全異於人類認知的方式來思考。這兒有一個關於非人類思考的一個很好的例子，今年三月在德克薩斯州奧斯汀舉行的西南偏南音樂節（South by Southwest festival）上，沃森電腦就上演了一幕厲害的絕技：IBM的研究員給沃森添加了由在線菜譜、美國農業部（USDA）出具的營養表以及讓飯菜更美味的味道研究報告組成的數據庫。憑藉這些數據，沃森依靠味道配置資料和現有菜色模型創造出了新式的菜餚。其中一款由沃森創造出的受人追捧的菜餚是美味版本的“炸魚和炸薯條”（fish and chips），它是用酸橘汁醃魚和油炸芭蕉製成。在約克城高地的IBM實驗室裏，我享用了這道菜，也吃了另一款由沃森創造出的美味菜餚：瑞士/泰式蘆筍乳蛋餅。味道挺不錯！

Nonhuman intelligence is not a bug, it's a feature. The chief virtue of AIs will be their alien intelligence. An AI will think about food differently than any chef, allowing us to think about food differently. Or to think about manufacturing materials differently. Or clothes. Or financial derivatives. Or any branch of science and art. The alienness of artificial intelligence will become more valuable to us than its speed or power.

非人類的智能不是錯誤，而是一種特徵。人工智能的主要優點就是它們的“相異智能”（alien intelligence）。一種人工智能產品在思考食物方面與任何的大廚都不相同，這也能讓我們以不同的方式看待食物，或者是以不同的方式來考慮製造物料、衣服、金融衍生工具或是任意門類的科學和藝術。相較於人工智能的速度或者力量來說，它的相異性對我們更有價值。

As it does, it will help us better understand what we mean by intelligence in the first place. In the past, we would have said only a superintelligent AI could drive a car, or beat a human at Jeopardy! or chess. But once AI did each of those things, we considered that achievement obviously mechanical and hardly worth the label of true intelligence. Every success in AI redefines it.

實際上，人工智能將幫助我們更好地理解我們起初所說的智能的意思。過去，我們可能會說只有那種超級聰明的人工智能產品才能開車，或是在“危險邊緣”節目以及國際象棋大賽中戰勝人類。而一旦人工智能做到了那些事情，我們就會覺得這些成就明顯機械又刻板，並不能夠被稱爲真正意義上的智能。人工智能的每次成功，都是在重新定義自己。

But we haven't just been redefining what we mean by AI—we've been redefining what it means to be human. Over the past 60 years, as mechanical processes have replicated behaviors and talents we thought were unique to humans, we've had to change our minds about what sets us apart. As we invent more species of AI, we will be forced to surrender more of what is supposedly unique about humans. We'll spend the next decade—indeed, perhaps the next century—in a permanent identity crisis, constantly asking ourselves what humans are for. In the grandest irony of all, the greatest benefit of an everyday, utilitarian AI will not be increased productivity or an economics of abundance or a new way of doing science—although all those will happen. The greatest benefit of the arrival of artificial intelligence is that AIs will help define humanity. We need AIs to tell us who we are.

但我們不僅僅是在一直重新定義人工智能的意義——也是在重新定義人類的意義。過去60年間，機械加工複製了我們曾認爲是人類所獨有的行爲和才能，我們不得不改變關於人機之間區別的觀點。隨着我們發明出越來越多種類的人工智能產品，我們將不得不放棄更多被視爲人類所獨有能力的觀點。在接下來的十年裏——甚至，在接下來的一個世紀裏——我們將處於一場曠日持久的身份危機（identity crisis）中，並不斷捫心自問人類的意義。在這之中最爲諷刺的是，我們每日接觸的實用性人工智能產品所帶來的最大益處，不在於提高產能、擴充經濟或是帶來一種新的科研方式——儘管這些都會發生。人工智能的最大益處在於，它將幫助我們定義人類。我們需要人工智能來告訴我們，我們究竟是誰。