李开复：OMO将引发中国经济新风暴

本文转载自微信公众号“创新工场”（ID：chuangxin2009），36氪经授权转载。

“未来世界即将迎来‘OMO’（Online-Merge-Offline线上与线下融合）的时代，而中国的发展速度相当惊人，将有望在全球范围内首先实现OMO。”

11月22日，创新工场创始人兼首席执行官李开复博士，在“经济学人”（The Economist）杂志的 The World in 2018 特辑发表专栏文章，结合出行、零售、教育等领域，描述了OMO给生活带来的改变。 

在过去三年，移动支付在中国掀起了一场巨大的风暴，培育了摩拜、滴滴、美团等公司。之后，人工智能技术赋能的传感器、计算机视觉感知逐步穿透实体世界，将把汽车、商店、商场、诊所和学校等现实世界的场景和行为实时数据化，自此，数据世界和实体世界将完全整合。

当然，在从中获益、享受便利的同时，我们还需要在这个全新的世界中保护个人隐私和安全。

两年前，早上当我从北京的办公室里往下看，看到的是车水马龙，拥堵非常严重。而今天我再往下看的时候，却是来去匆匆的橘黄色自行车，这就是现在的上班族骑的“智能”自行车。如今， “智能”自行车正在悄悄改变中国繁华城区早晚高峰时的景象。 

“智能”自行车的代表品牌——摩拜单车，是创新工场所投资的一家企业，它在不到一年的时间里实现了每天2500万的骑行量。这些都在显示，在即将到来的2018年，我们将迎来一场有力的变革：线上与线下的界限被消除。

这些单车并不是北京早期黑白照片中那些生锈破旧的自行车，而是装上了太阳能全球定位系统、加速器、蓝牙和热探测器的智能体。用户通过智能手机，就能激活车上的近场通讯和麦克风。注册骑行也很简单，用手机支付押金后，摩拜单车上的智能锁就会自动打开。

在骑行过程中，各种传感器会把用户的移动坐标和其他数据传输到云计算服务器。每天，中国大城市的数百万摩拜单车会产生20T的数据，并反馈到云服务器上，从而将人、自行车、道路和目的地连接起来，构成了全球最大的“物联网”网络。摩拜单车服务器采用人工智能来分析交通状况并平衡供需关系，从而实现效率最大化。

我把这种现象称之为即将到来的未来世界“OMO”（线上与线下融合）。四个因素将助力OMO时代的到来：智能手机的大规模应用、流畅的支付系统、质优价廉的传感器、以及人工智能技术的进步。在所有这些领域，中国的发展速度相当惊人，并且未来将有望首先实现OMO。 

中国现在有7.31亿智能手机用户，这些智能手机上都配置有全球定位系统和传感器。在这7.31亿用户当中，70%的人在购物时，会使用移动支付取代信用卡或者现金付款。流畅的电子付款几乎是实时到账，通常不收取手续费，而且不限制最低购买量。此外，不像银行卡只能用于个人对企业转账，移动支付可以实现个人对个人转账。

在过去的三年里，这种购物方式在中国的经济领域掀起了一场巨大的风暴。除摩拜单车外，便捷的移动支付系统还为滴滴、美团和许多其他新兴企业铺平了道路。其中滴滴相当于中国的优步，美团每天有1000万的外卖订单量。现在这些都在朝着国际化的方向发展。

在未来，我们日常使用的工具中将更普遍地安装传感器，从而产生更多的数据。这些传感器可以捕捉个人地理位置、动向等信息，并将这些信息传输到服务器。这些信息与在线数据相结合，通过人工智能技术的分析，将为众多行业的发展带来一场变革。

F5未来商店也是创新工场投资的一家企业。这是一家自动化商店，总部位于广州。目前，“自动化商店”正在呈现另一片OMO发展蓝图。F5未来商店出售普通便利店中的各种热卖单品，包括热餐。

跟亚马逊无人超市Amazon Go的模式类似，F5未来商店从备餐到结账都通过机器实现了全程自动化，但F5抢先一步，首先实现了规模化和盈利。未来，这些商店都会装上传感器，识别客户的身份、动向、行为甚至是意图，就像用户浏览电子商务网站信息一样，实现无缝跟踪。 

最后一个例子是教育行业。语言学习在中国十分盛行，未来的语言学习将会把外籍老师、本地助教、自动化软件和自动化硬件连接起来：由外籍老师负责远程授课，本地助教负责活跃气氛，自动化软件负责纠正发音，自动化硬件负责作业和测试评分。通过对VIPKID、盒子鱼以及七天教育的投资，我们已经帮助他们在中国的学校和培训中心实现了这种模式。

一旦在汽车、商店、商场、诊所和学校里安装更多的传感器，那些能获取数据的企业就能够知道并追踪消费者的行为，甚至获取的数据比通过跟踪在线信息获取的还要多。

通过这种方式，就可以知道一个人选了和买了哪些东西，或者又把哪些商品放回了货架；并能够获知用户是否感冒发烧或者是否在楼梯间摔倒；以及他们要去哪里，并推断他们可能会做什么（例如午餐会议、就诊、回家等）。

下一步，服务方将会实现线下与线上数据融合，提供精准的客户推荐，提升店内服务，并实现供应链自动化和即时库存管理。 

未来OMO和人工智能将推动消除线上与线下数据的区分，我们能够从中获得巨大的经济利益，享受前所未有的便利。但与此同时，我们也需要在这个全新的融合世界中保护个人隐私和安全。

本文英文版刊于The Economist出版的The World in 2018。《财经》杂志将在2018年2月独家出版The World in 2018中文版《世界2018》。

Meet OMO sapiens

Two years ago, the morning view from my office in northern Beijing was a snaking traffic jam of cars. Today I look down to see a rush of orange and yellow bicycles, as commuters have leapt onto the “smart” bicycles that are reshaping rush hour in China’s busiest cities. In less than a year Mobike (in which my venture fund invested ) has reached 25m rides per day. That illustrates another force that will be vastly more transformative in 2018: the boundaries between the online and offline worlds are being erased.

These bikes are not the rusty clunkers you may have seen in old black-and-white photos of China’s capital. They are equipped with solar-powered GPS, accelerator, Bluetooth and a thermos detector. Both NFC (near-field communications) and microphones are activated by smartphone. To sign up to ride, you make a digital payment from your phone, and the bike’s smart lock automatically opens.

As you cycle around the city, the various sensors transmit your moving co-ordinates and other data to a cloud-based server. Each day, millions of cyclists peddling around Chinese megacities generate 20 terabytes of data, which feed back into Mobike’s cloud servers— connecting people, bikes, roads and destinations as one of the world’s largest “internet of things” networks. Mobike’s servers use artificial intelligence (AI) to analyse traffic patterns and to balance supply and demand, maximising usage and alerting the company to any problems.

I call this futuristic but imminent world “OMO” (online merges with offline). Four factors are enabling the arrival of OMO: rapid smartphone uptake, frictionless payment systems, cheaper and better sensors, and advances in AI. In each of these areas, China is moving extraordinarily fast, and is poised to see the OMO future first.

China now has 731m people wielding smartphones, all equipped with GPS and other sensors. Some 70% of them use their phones for digital payments, in lieu of credit card or cash. This frictionless digital payment is nearly instantaneous, usually zero-fee, requires no minimum purchase and can be used person-to-person (not only person-to- business, like credit cards). 

In the past three years this new way of buying things has taken the Chinese economy by storm. Easy digital payment systems have paved the way for not just Mobike, but also Didi (the Uber of China), Meituan (delivering over 10m take-out meals per day) and many other new companies. Many of these are now also expanding internationally.

In the year ahead, more sensors will be added to more tools of daily life, all generating more data. “Facts”, such as a person’s location, movements and even identity, are beginning to be captured by these sensors and transmitted online. The information will be combined with online data and analysed by AI, transforming the future of many industries.

 “Autonomous stores”, such as the F5 Future Store (in which my fund also invested) with headquarters in Guangzhou, are devoloping another blueprint for OMO. F5 sells popular items common in regular convenience stores, including hot meals. Like Amazon Go, F5 is fully autonomous (done by machine), from meal preparation to check-out. But in a step ahead of Amazon Go, F5 is scalable and profitable. In the future, these stores will be equipped with sensors that can discern a customer’s identity, movements, behaviour and even intent, as seamlessly as if someone were clicking around an e-commerce website.

A final example is in education. Language learning, which is extremely popular in China, will combine native-speaking teachers lecturing remotely, local assistants keeping the atmosphere fun, autonomous software correcting pronunciation and autonomous hardware grading homework and tests. Our investments in VIPKID, Boxfish and Septnet have already built capabilities into schools and training centres in China.

Brave new world

Once there are more sensors in cars, stores, malls, clinics and schools, those with access to the data will know and track each person’s behaviour even better than when that person is online. It will be possible to know what product someone picked up and bought, or put back on a store shelf; whether they are running a fever or just tripped on a stairwell; and where they went and, by inference, what they probably did (eg, a lunch meeting, a hospital visit, returned home). As a next step, offline and online data can be combined, providing accurate customer recommendations, improving in-store services, automating the supply chain and achieving just-in-time inventory management.

OMO and AI will take us into a future where any distinction between these worlds disappears. We will reap great financial benefits, and enjoy unprecedented convenience, but we also need to find ways to protect people’s privacy and safety in this brave new world.

The online and offline worlds will merge, predicts Kai-Fu Lee, CEO of Sinovation Ventures and president of Sinovation Ventures Artificial Intelligence Institute.