Ali's Smart City Platform Suffers from Uniquely Chinese flaws
Behind the stories of international exports and privacy concerns, the City Brain reveals a great deal about the political system to which is belongs.
By Luke Cavanaugh, who produces interweave.gov, a newsletter covering digital government developments in Europe and Asia.
In early 2018, Alibaba Cloud executive Min Wanli addressed a packed conference center in Kuala Lumpur. “Every moment [human beings] are sensing different signals either through our eyes, our hearing, our speech, our text messages,” he began. “The body digests and extracts the most important information and then decides what action you want to take.” But why, he asked, could a city not do this too?
Min’s question was far from hypothetical. He was discussing the technical detail of a collaboration between the Malaysia Digital Economy Corporation (MDEC) and Alibaba on the introduction of the City Brain to Malaysia. It was a project that Min had been working on in Hangzhou, where Alibaba is headquartered.
The City Brain began in 2016 as an intelligent transport system. Its artificial intelligence focused on processing data in real time based on current traffic status and speed information, using it to adjust bus frequencies and control traffic lights. Hangzhou, once one of the most congested cities in China, saw its average congestion time reduced by more than 15% during the platform’s first year of operation.
Its scope today is far broader. From the Industrial Brain to the Medical Brain, the Environmental Brain to the Financial Brain, Alibaba is is deploying artificial intelligence and neural networks across Chinese society.
But as a standout example among China’s Smart Cities, the City Brain is the product of a unique political system. Government backing has allowed the City Brain unparalleled access to data and the ability to scale. With total Chinese investments in Smart Cities surpassing 2.4 trillion yuan in 2020, and Xi Jinping himself reiterating the importance of China’s digital government strategy this year, the stakes for success are high.
And it’s not just the ongoing tension between Alibaba and the government that threatens to derail its vision.
ChinaTalk is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
The City Brain’s “Version 3.0” expanded its scope beyond congestion, introducing functions to prepare the city for natural disasters and even streamline the city’s tourism. Its launch brought more than just improved City Brain functionality. It also reinforced the global ambition of one of China’s biggest companies.
Speaking at the launch of City Brain 3.0, Alibaba Cloud President Zhang Jianfeng announced:
“A hundred years ago, London exported the subway to the world, Paris exported the sewer, and New York exported the power grid. Today, the value of digital leadership in Chinese cities has been highlighted, and the establishment of a new digital foundation has become the foundation for the evolution of future cities.”
China’s export of smart city models has significant ramifications for the rest of the world, not least in the controversies surrounding the City Brain’s use of surveillance.
There is scrutiny in the West over the predominance of illiberal regimes placing orders for Chinese Smart City solutions. Among them, Uganda was accused of using Huawei facial recognition technology to track down dissidents following anti-government protests in 2020.
A Washington-based advisory company found that of the 64 countries that had signed up to install Chinese smart city technology, 41 were ranked as “not free” or “partly free” by Freedom House.
Concerns also proliferate around Chinese companies having access to sensitive data. With Beijing’s National Security Law requiring companies to comply with Chinese government directives, including data handover demands, this fear has grown in recent years.
But this is nothing new. As far back as 2018, French newspaper Le Monde reported that Huawei had installed listening devices in the headquarters of the African Union in Addis Ababa, Ethiopia. The company had provided most of the computer systems for the building, which appeared to be transferring data to China each evening.
While these concerns over using Chinese smart city technologies abroad are well documented, less so is what the City Brain tells us about China’s internal politics and the extent to which they are compatible with the rest of the world.
Much of the success of the City Brain is down to two unique features of China’s political ecosystem: its decentralized and experimental policymaking environment and its close public-private relationships. But these same relationships that provide the impetus for innovation have ultimately also led to over-centralization, stifling the entry into the market of new players and threatening the sector with oligopoly.
A brain of three parts
The metaphor of the brain fits Alibaba’s Smart City solution, which is ostensibly made up of two parts. First, there is the “nervous system,” a network of sensors across Hangzhou, gathering data from traffic lights and street cameras. Then there is the “brain” itself, a central artificial intelligence system that runs on Aspara, Alibaba’s computing operating system. The City Brain gathers information, aggregates it and uses it to optimize the city’s management.
To monitor Hangzhou’s roads, the City Brain draws on data from videos, wi-fi probes, carriers, and other sources to track passenger delay rates and estimate capacity needs.
It uses cloud computing to process this data in real time, allowing the brain to adjust bus frequencies based on travel supply and demand, monitor taxi dispatches to minimize delay rates at key venues and transport hubs, and control traffic lights so that ambulances can pass through without running any red lights.
The third part of the City Brain, China’s political ecosystem itself, is less tangible. Digital Government has been a focus in Chinese policy documents stretching back to the era of the 10th (2001-2005) and 11th Five-Year plans (2006-2010). In broad strokes, policy documents focus on governing a population of 1.37 billion in a timely and cost-efficient manner.
But precise policy direction has often been vague. The 12th Five-Year Plan made it an official strategy that local, prefecture and national governments focus on developing digital government.
The 14th and most recent five-year plan stated that “innovation must be placed at the heart of China’s development and advanced in every field, from theory to institutions, science, technology and culture.” However, it did little to clarify what this might look like in the context of digital government.
Consequently, the responsibility for specific policy direction is often borne by provincial governments themselves. To this end, the City Brain has arisen out of China’s unique model of local experimental policy making. This process involves using local governments as incubators, whereby initiatives can be trialled at a local level to solve a national issue. If successful, they are then implemented nationally.
This type of policy experimentation has become commonplace in China. Beyond the realm of digital government, China’s rural pension system was developed after some 2,000 rural counties took part in a trial of different social pension systems. The most successful trial, that of Baoji City, was then rolled out across the country.
The implications of this system are threefold. Firstly, it requires that policy is innovated and implemented at a local level to facilitate comparison between different options.
Secondly, it allows Chinese local governments to fail fast. For every policy option that gets adopted, thousands do not.
Thirdly, and most importantly concerning the City Brain, it lends itself to strict top-down control over policies from local governments. To roll out a policy across a country, it is necessary that it can be replicated and that as many variables as possible are controlled.
Alibaba’s brain technology took less than three years to go from the first implementation in Hangzhou to placement in 22 cities across China.
China’s promotion system for local political cadres complements this culture of controlled experimentation. Within the Chinese political system, various localities are treated as autonomous or independent units, allowing for “yardstick competition among local officials.” As Li and Zhou show in a 2005 study, local government politicians (including those at a prefecture level) are often promoted into the national government based on their region’s economic performance.
Pressure on local governments to innovate, then, comes both from the central government and from other innovative local governments. These circumstances gave rise to the City Brain, its continual innovations, and its replication across China.
A relationship of multidimensional participation
To focus on innovation within Hangzhou’s local government would be to only tell half the story. The City Brain is ultimately a solution developed by the private sector. But Alibaba’s relationship with authorities in Zhejiang is just as important as the innovative potential of the local governments themselves.
In an interview about the role of Smart Cities in China, the Head of Government Services at Alibaba’s Cloud Smart Digital Division, Xu Shijun was asked about the company’s relationship with the government:
“Alibaba has proposed a deep partnership, not only in business but also in capital cooperation. We’re advancing from the previous model of constructing a single project to today’s model of multidimensional participation in planning, design, construction, service and operation. All round participation.”
A month later, in September 2019, this relationship was to become closer still. Hangzhou authorities placed government affairs representatives in 100 of Hangzhou’s biggest firms, including Alibaba.
Alibaba originally pitched the City Brain as a solution to a tender put out by Hangzhou’s government to solve the problem of congestion in the city. Since then, the company has collaborated with the city on a number of digital government projects.
At the beginning of the Covid-19 pandemic, the city authorities put out a tender to contract China’s first Public Service and Management Platform (PSMP) for the prevention and control of the pandemic. With Alibaba’s technological prowess and existing relationship with the government, they completed the platform’s design in just one day.
A couple of weeks later, the municipal government requested that Alibaba develop a smartphone-based health code program for employees returning to work in the city. Research on the app began on Feb 6, 2020; five days later it was ready for testing, and by Feb 12 it launched.
A victim of its own design?
All of a sudden, amid a controversy surrounding its founder Jack Ma, Alibaba was hit with a record 18 billion yuan anti-trust fine by China’s central government.
At the same time, Zhejiang became the first province to issue competition guidelines against online platform operators. Earlier this year, the province’s Market Supervision Bureau also announced further steps to promote a healthy business environment, including implementing a credit classification and setting up a fair competition review and anti-monopoly committee.
These anti-trust actions exposed the risk for the government of depending on one major supplier to produce a digital government platform. Alibaba is not the only player in the City Brain. Hikvision — China’s controversial CCTV manufacturer — provides much of the hardware within the City Brain, for example. But Alibaba remains by some distance the largest private company involved in the platform.
The risks of the top-down approach that enabled the City Brain’s rapid growth run deeper still. As UK Public Purpose Technology expert Tanya Filer notes, “tendering a specific solution that is pre-defined internally, as public sector agencies tend to do, risks excluding from the process start-ups that have high-quality and viable products and services with strong product-market fit.”
By continually depending on the speed and scale of Alibaba to meet proscriptive tenders, and strengthening its influence over the company with government affairs representatives, the Zhejiang government has excluded from the ideation process non-government voices, particularly start-ups.
The result is a project that depends on government policy and creativity to drive its innovation. Earlier this year, WEF founder Klaus Schwab and economist Thierry Malleret published a companion article to their recently released book The Great Narrative. In it, they reflect on the phrase “failure of imagination,” associated “with situations in which strategic thinking and risk management are stuck in unimaginative and reactive thinking:”
“Imagination is precisely what is required to escape a state of “cognitive lockdown” and to build a “cohesive whole.” It gives us the capacity to dream up innovative solutions to successfully address the multitude of risks that confront us.”
A cohesive whole is exactly what the City Brain was designed for: a comprehensive management platform for a city. But imaginative solutions work best when they are inclusive.
Those start-ups excluded from the tendering process for the City Brain may have designed solutions to enhance China’s digital government, or even have provided better alternatives. They may not have done. But by failing to invest in championing voices other than large “national champions” and the government itself, Hangzhou’s authorities are failing to profit from the imaginative potential that independent business accelerators, hackathons, market intelligence, policy actors and start-ups provide.
In restricting its opportunities to benefit from independent imaginative thought, those that run the City Brain may just have created a system that is a victim of its own design.
ChinaTalk is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
Thanks for a great story.