李开复| 人工智能对人类社会的真正威胁
日前，创新工场创始人兼CEO、人工智能工程院院长李开复博士在纽约时报（The New York Times）的“观点”（Opinion）栏目发表专栏文章《人工智能对人类社会的真正威胁》（The Real Threats of Artificial Intelligence）。
在人工智能已经深入生活的今天，社会上不乏“人工智能威胁论”，担忧机器人会“反噬”人类。在这篇文章里，李开复博士讨论了人工智能技术未来发展所带来的几个更真切和亟待解决的问题：全球性的失业问题及可能产生的全球性经济失衡和贫富差距。
在2011年纽约时报的一篇专栏文章中，李开复博士曾预测中国手机支付直接跳过信用卡取代现金而成为主流支付方式、线上教育会进入二三线城市、本地餐饮使用发生颠覆。当时，微信还没有出现，而如今我们回顾这三个预测，基本都已经发生。
因此我们有理由相信，这次的预测同样并非科幻小说式的想象，文中的观点值得深思。
以下为专栏文章的完整中译版及英文原文。
中文版
面对呼之欲出的人工智能时代，您最担心什么？
通常，人们对于这个问题的回答很像各类科幻片中的惊悚情节。他们担心人工智能的发展会带来所谓的“奇点”，即在人类发展的某一特定历史时刻，人工智能会完全超越人类智慧，继而将人类社会带入一场无法想象的变革当中。人们甚至开始怀疑，人工智能是否最终会控制人类，使人类沦为所谓的“机械人”。
这些问题值得探讨，但并非亟待解决。先不论这些问题是否会发生，即使哪天真的出现，也是数百年以后。而目前，人类还没有任何已知的途径和方法能够将当前最卓越的人工智能系统——比如刚刚战胜了最出色的人类棋手柯洁的围棋计算机程序AlphaGo，转化为通用的人工智能，即具有自我意识、可进行常识性推理、能够自觉地从多领域获取知识、并具有感知、表达和理解等能力的电脑程序。
但这并不意味着我们就可以高枕无忧。恰恰相反，现有人工智能技术和产品的发展速度之快大大超出我们的认识和预期，人工智能技术注定会改变我们的世界，并不完全以我们的意愿为转移。人工智能是工具，不是一种智慧形式。但它注定会重新定义工作的意义以及财富的创造方式；值得注意的是，它将带来前所未有的经济失衡现象，甚至改变全球的权力格局。
因此，当务之急，让我们先对这些迫在眉睫的现实挑战予以关注。
人工智能到底是什么？粗略来讲，人工智能技术指的是获取某一领域（比如贷款偿还记录）的海量信息，并利用这些信息对具体案例（是否应给某人贷款）做出判断，以达成某一特定目标（贷方利益最大化）的技术。这些技术在给定任务中所展现出的工作能力已经被证明可以完全超越人类的表现。
今天，这样的人工智能技术正在被广泛应用于各个领域。随着它的进一步发展，会不可避免地对就业造成冲击。很多岗位和职业会逐步消失，例如银行出纳员、客户服务代表、电话销售员、股票和债券交易员等；甚至律师助理和放射科医生这样的工作也会被这类软件所取代。假以时日，人工智能技术还会学会控制如无人驾驶汽车和机器人这类半自主或全自主硬件设施，逐步取代工厂工人、建筑工人、司机、快递及许多其他职业。
与工业革命及信息革命不同，人工智能技术所带来的冲击并非单纯指向某些特定岗位和职业，如传统制造业中的手工艺者被流水线工人所取代；或只会使用纸张和打字机的秘书被精通电脑的个人助理所替代等；人工智能所带来的是对现有职业和工作版图大规模地颠覆。毋庸讳言，其中大部分为低薪工作，但某些高薪岗位也将面临挑战。
值得注意的是，这场变革将会为开发人工智能技术及采用人工智能技术的公司和企业带来巨额利润。试想，如果优步能全面利用无人驾驶车进行运营；苹果公司能够省却大量人力生产其产品；全年满足超过三千万笔贷款请求却不需要任何人工干预的借贷公司；可以想见，这些企业将利用人工智能技术创造何等惊人的利润和收益！而这一切已经是现在进行时。创新工场最近就在国内投资支持了一家利用人工智能技术进行借贷的的初创企业。
诚如你所看到的，人类正面临着很难妥善共存的两个发展前景：一方面我们迎来了仅用少量人力就能创造巨大财富的发展时代，而另一方面，大量人员也将因此而下岗和失业。各种权衡，何去何从？
答案之一当然是教育，即要对人工智能所不擅长的领域进行有针对性的人员教育和再培训。具体来说，人工智能并不擅长需要创造力、规划能力以及“跨领域”思考能力等类型的工作——比如辩护律师。这些能力也是目前很多高端职位所要求的，问题是通过短期培训来传授和获取这些能力和技能的可能行较低。另一个方向则是弥补人工智能系统所欠缺的“人际交往能力”，发展出更多类似社会工作者、酒保、按摩技师等需要人际间微妙互动的岗位。即便如此，另一个问题随之出现：我们的社会对酒保或类似岗位又有多大需求呢？
按照我的个人推测，要解决人工智能变革所带来的大规模失业问题，需要的是更多我所说的所谓“关爱服务”。 这是人工智能无法完成，而社会又大量需要的服务；更不用讲你我生而为人所赖以的使命感和荣誉感。此类服务岗位不胜枚举，例如：陪伴老人就医的志工、孤儿院的教导员、戒酒互助社的志愿者，甚或未来可能出现的——帮助那些沉迷于电脑虚拟现实刺激中的“平行人”重返人生现实的热心人。换言之，当下的很多所谓志愿服务工作未来都可能成为真正的职业。
其中一些服务甚至会转变为高薪职业并趋于专业化，例如可协助和配合“人工智能癌症诊断程序”工作的、具有专业医疗知识、同时又富有同情心和极强沟通技巧的医疗服务提供者。总体而言，人们可以选择比现在更短的工作时间。
那么，谁会为这些工作买单呢？文章开始时我提到的那些集中于相对少数企业手中的巨额财富现在可以派上用场了。在我看来，人工智能所创造财富中的相当一部分会不可避免的转移到那些工作被取代了的人们那里去。而这一过程似乎只能是通过凯恩斯主义的财政政策——即提高政府相关领域的开销，及增加高利润公司的税收来加以实现。
至于那样状况下的社会福利是何种形式，我认为可能是一种有条件的全民基本收入方案，即社会福利将面向有经济需求并符合条件的人群。所谓“条件”，是指福利申请者必须努力参与就业或再就业培训，或保证参与一定工时的“关爱服务”。
当然，为了给这类社会福利提供资金，提高税率可能在所难免。政府不仅要补贴大部分人的生活和工作，还要设法对此前大量下岗员工无法缴纳的个人所得税进行弥补。
这就带来了关于人工智能最终、也是最重要的挑战。我所描绘的凯恩斯主义的财政政策或许在美国和中国是可行的，因为这两个国家可以通过其规模巨大且成功的人工智能企业来获取税收，并以此支撑其高昂的社会福利方案。但是其它国家又当如何呢？
相较而言，其他国家会面临两个难以克服的问题。首先，大部分人工智能所创造的财富会流入美国和中国。人工智能是一个“强者更强”的产业：数据越多，产品越好；产品越好，所能获得的数据就更多；数据更多，就更吸引人才；人才越多，产品就会更好。在这个良性循环里，中美两国目前已经汇聚了大量人才、市场份额以及能够调动的数据。
举例来说，中国的语音识别企业科大讯飞以及人脸识别公司如旷视科技、商汤科技等就市值来讲，都已经成为行业翘楚。在谷歌、特斯拉及优步等企业的引领下，美国的无人驾驶技术也是首屈一指。而在消费互联网领域，中美七家企业——谷歌、脸书、微软、亚马逊、百度、阿里巴巴、腾讯——都已在其现有产品和服务中大量使用人工智能技术，并正快速将其运营版图扩展到全球范围内，尽可能占据更大份额的人工智能市场。从目前的情势看，美国似乎占据发达国家市场及部分发展中国家市场，而中国公司无疑赢得了多数发展中国家市场。
对于中国和美国以外的其他国家来讲，另外一项挑战则在于许多国家还在日益增长的人口，尤其是一些发展中国家。庞大的人口可以成为一种经济资本，就如同其近几十年来在中国和印度的经济发展中所产生的积极作用。但是在人工智能时代，这一资本却可能成为经济负担，因为其中大部分人口将面临下岗失业。
所以，如果很多国家不能通过向高额盈利的人工智能企业征税来补贴工人，他们还能有什么其他选择？依我个人推论，为避免本国人民陷入贫困，这些国家会与提供最多人工智能软件的国家——中国或者美国——进行磋商和谈判，最后以特定人工智能企业在本地用户中的盈利来换取国家所需的社会福利补贴。从而最终成为中美两国的经济依附体，这样的经济发展态势也将重塑当今的地缘政治版图。
一言以蔽之，最大程度地缩小人工智能可能造成的经济失衡和贫富差距，已是当下必须要考虑的问题，此差距不仅体现在国家内部，也体现在国与国之间。从乐观的角度看：人工智能为我们展现了一个打破全球经济失衡状态的机会，而挑战所带来的巨大影响，将使任何国家都无法置身事外。
英文版
What worries you about the coming world of artificial intelligence?
Too often the answer to this question resembles the plot of a sci-fi thriller. People worry that developments in A.I. will bring about the “singularity” — that point in history when A.I. surpasses human intelligence, leading to an unimaginable revolution in human affairs. Or they wonder whether instead of our controlling artificial intelligence, it will control us, turning us, in effect, into cyborgs.
These are interesting issues to contemplate, but they are not pressing. They concern situations that may not arise for hundreds of years, if ever. At the moment, there is no known path from our best A.I. tools (like the Google computer program that recently beat the world’s best player of the game of Go) to “general” A.I. — self-aware computer programs that can engage in common-sense reasoning, attain knowledge in multiple domains, feel, express and understand emotions and so on.
This doesn’t mean we have nothing to worry about. On the contrary, the A.I. products that now exist are improving faster than most people realize and promise to radically transform our world, not always for the better. They are only tools, not a competing form of intelligence. But they will reshape what work means and how wealth is created, leading to unprecedented economic inequalities and even altering the global balance of power.
It is imperative that we turn our attention to these imminent challenges.
What is artificial intelligence today? Roughly speaking, it’s technology that takes in huge amounts of information from a specific domain (say, loan repayment histories) and uses it to make a decision in a specific case (whether to give an individual a loan) in the service of a specified goal (maximizing profits for the lender). Think of a spreadsheet on steroids, trained on big data. These tools can outperform human beings at a given task.
This kind of A.I. is spreading to thousands of domains (not just loans), and as it does, it will eliminate many jobs. Bank tellers, customer service representatives, telemarketers, stock and bond traders, even paralegals and radiologists will gradually be replaced by such software. Over time this technology will come to control semiautonomous and autonomous hardware like self-driving cars and robots, displacing factory workers, construction workers, drivers, delivery workers and many others.
Unlike the Industrial Revolution and the computer revolution, the A.I. revolution is not taking certain jobs (artisans, personal assistants who use paper and typewriters) and replacing them with other jobs (assembly-line workers, personal assistants conversant with computers). Instead, it is poised to bring about a wide-scale decimation of jobs — mostly lower-paying jobs, but some higher-paying ones, too.
This transformation will result in enormous profits for the companies that develop A.I., as well as for the companies that adopt it. Imagine how much money a company like Uber would make if it used only robot drivers. Imagine the profits if Apple could manufacture its products without human labor. Imagine the gains to a loan company that could issue 30 million loans a year with virtually no human involvement. (As it happens, my venture capital firm has invested in just such a loan company.)
We are thus facing two developments that do not sit easily together: enormous wealth concentrated in relatively few hands and enormous numbers of people out of work. What is to be done?
Part of the answer will involve educating or retraining people in tasks A.I. tools aren’t good at. Artificial intelligence is poorly suited for jobs involving creativity, planning and “cross-domain” thinking — for example, the work of a trial lawyer. But these skills are typically required by high-paying jobs that may be hard to retrain displaced workers to do. More promising are lower-paying jobs involving the “people skills” that A.I. lacks: social workers, bartenders, concierges — professions requiring nuanced human interaction. But here, too, there is a problem: How many bartenders does a society really need?
The solution to the problem of mass unemployment, I suspect, will involve “service jobs of love.” These are jobs that A.I. cannot do, that society needs and that give people a sense of purpose. Examples include accompanying an older person to visit a doctor, mentoring at an orphanage and serving as a sponsor at Alcoholics Anonymous — or, potentially soon, Virtual Reality Anonymous (for those addicted to their parallel lives in computer-generated simulations). The volunteer service jobs of today, in other words, may turn into the real jobs of the future.
Other volunteer jobs may be higher-paying and professional, such as compassionate medical service providers who serve as the “human interface” for A.I. programs that diagnose cancer. In all cases, people will be able to choose to work fewer hours than they do now.
Who will pay for these jobs? Here is where the enormous wealth concentrated in relatively few hands comes in. It strikes me as unavoidable that large chunks of the money created by A.I. will have to be transferred to those whose jobs have been displaced. This seems feasible only through Keynesian policies of increased government spending, presumably raised through taxation on wealthy companies.
As for what form that social welfare would take, I would argue for a conditional universal basic income: welfare offered to those who have a financial need, on the condition they either show an effort to receive training that would make them employable or commit to a certain number of hours of “service of love” voluntarism.
To fund this, tax rates will have to be high. The government will not only have to subsidize most people’s lives and work; it will also have to compensate for the loss of individual tax revenue previously collected from employed individuals.
This leads to the final and perhaps most consequential challenge of A.I. The Keynesian approach I have sketched out may be feasible in the United States and China, which will have enough successful A.I. businesses to fund welfare initiatives via taxes. But what about other countries?
They face two insurmountable problems. First, most of the money being made from artificial intelligence will go to the United States and China. A.I. is an industry in which strength begets strength: The more data you have, the better your product; the better your product, the more data you can collect; the more data you can collect, the more talent you can attract; the more talent you can attract, the better your product. It’s a virtuous circle, and the United States and China have already amassed the talent, market share and data to set it in motion.
For example, the Chinese speech-recognition company iFlytek and several Chinese face-recognition companies such as Megvii and SenseTime have become industry leaders, as measured by market capitalization. The United States is spearheading the development of autonomous vehicles, led by companies like Google, Tesla and Uber. As for the consumer internet market, seven American or Chinese companies — Google, Facebook, Microsoft, Amazon, Baidu, Alibaba and Tencent — are making extensive use of A.I. and expanding operations to other countries, essentially owning those A.I. markets. It seems American businesses will dominate in developed markets and some developing markets, while Chinese companies will win in most developing markets.
The other challenge for many countries that are not China or the United States is that their populations are increasing, especially in the developing world. While a large, growing population can be an economic asset (as in China and India in recent decades), in the age of A.I. it will be an economic liability because it will comprise mostly displaced workers, not productive ones.
So if most countries will not be able to tax ultra-profitable A.I. companies to subsidize their workers, what options will they have? I foresee only one: Unless they wish to plunge their people into poverty, they will be forced to negotiate with whichever country supplies most of their A.I. software — China or the United States — to essentially become that country’s economic dependent, taking in welfare subsidies in exchange for letting the “parent” nation’s A.I. companies continue to profit from the dependent country’s users. Such economic arrangements would reshape today’s geopolitical alliances.
One way or another, we are going to have to start thinking about how to minimize the looming A.I.-fueled gap between the haves and the have-nots, both within and between nations. Or to put the matter more optimistically: A.I. is presenting us with an opportunity to rethink economic inequality on a global scale. These challenges are too far-ranging in their effects for any nation to isolate itself from the rest of the world.
（编辑:xueqi）
京公网安备11010802022018号