Artificial intelligence (AI) is the superpower of this era—and the main answer to China’s pursuit of world domination. By 2030, the country wants to become the world’s largest hub for AI innovation, with the AI industry projected to hit a valuation of $202.57 billion by 2026. The biggest impact of this ambition is that it would shift the China-US domination game on to the technology front, with the US being the centre of AI research for decades—going back to the 1950s.
How does China plan to up its game? In 2017, Beijing announced that the future of AI will be made in China, following which the State Council laid out the New Generation Artificial Intelligence Development Plan, which testified the country’s preparation for various policies and multibillion-dollar investment initiatives to fund moonshot projects, startups and academic research. This plan was developed in response to China’s trade war with the US and to build a domestic industry worth $150 billion.
In the same year, the Semantic Scholar project which was developed by the Allen Institute for AI had analysed more than 2 million academic AI papers published through the end of 2018 — and it seems that China has already beaten the US in published AI papers. Arguably, research papers are a proxy to a country’s developments — but nevertheless it shows that China is predicted to take over the US based on citations in more than 50 percent of AI research papers in 2019, in 10 percent of the most-cited papers this year and in the 1 percent of most-cited papers by 2025.
For the uninitiated, China’s AI development plan can be broken down into three stages. The first stage aims to match China to the world’s leading AI powers by 2020. The second stage is focused on making the country the leader in some aspects of the technology by 2025 and the final stage will see China become the ‘world’s premier artificial intelligence innovation centre’ by 2030.
Preparing initiatives to build a robust AI ecosystem
In fact, the New Generation Artificial Intelligence Development Plan and another state-sponsored agenda Made in China 2025 are the driving forces behind China’s steady progress in AI technology. This is especially true because in 2017 both documents caught the attention of China’s leadership, including President Xi Jinping, who then led the Communist Party’s Politburo to conduct an in-depth study session about AI in the following year.
The session was focused on how to develop, control and use this technology to shape the country’s future directions in its quest to become the world leader in AI. The outcomes of such sessions are strategically reserved for high-priority policy issues which will enable Chinese leaders to make meaningful decisions with the help of industry expertise.
Xi in his speech during the study session said “AI is a vital driving force for a new round of technological revolution and industrial transformation, and accelerating AI development is a strategic issue to decide whether we can grasp opportunities.” In this context, the New Generation Artificial Intelligence Development Plan reinstates that China should ‘achieve world-leading levels’ in AI technology, while the Made in China 2025 agenda points out that the country should aim to significantly reduce ‘external [foreign] dependence for key technologies and advanced equipment.’
New developments in China’s AI innovation
Positively, China is on its way to making significant progress on the AI front. Last year, China hosted the 2019 World Artificial Intelligence Conference (WAIC) for the first time in Shanghai to promote AI innovation globally. More than 150 global leading figures from the AI industry and academic circles participated in the conference, along with 300 domestic and overseas enterprises.
But that was not the only reason why the WAIC was of great significance to China’s AI ambition. The conference saw IBM China sign a three-year agreement with Shanghai Zhangjiang Group to build the first Watson Build Artificial Intelligence Innovation Centre in the country. The idea to establish a holistic AI centre comes from Shanghai’s three-year action plan to build itself as an AI highland by 2021.
The main highlight of the proposed Watson Build Artificial Intelligence Innovation Centre is that it will be strategically located on the AI island of Shanghai’s Zhangjiang Science City — a hub for the AI industry — to promote AI talent and speed up AI applications. The centre is anticipated to be hugely beneficial to technology companies as it will potentially serve 500 domestic and foreign companies. Also, the centre will take a holistic approach to AI development by incubating between 30 to 60 intellectual property and patents and training 500 employees annually within the stipulated time.
Now, data shows that there are more than 1,000 core AI enterprises and more than 3,000 pan-AI enterprises established in Shanghai. The worth of AI-related industries is more than 70 billion yuan—pointing to the city’s genuine breakthroughs in this technology.
Contributing to these developments, China’s AI platform company AI ERA published a white paper last year on the country’s AI unicorns. According to the white paper, China had a total of 50 AI unicorns as of March 2019, with a total value of 3.5 trillion RMB. It appears that these unicorns are dominant in automotive transport and enterprise services. Statistically, 30 percent of those AI unicorns are dominant in automotive transport, while 26 percent of the total cover enterprise services. 28 out of 50 AI unicorns are located in Beijing in Haidian district, seven in Zhejiang, six in Shanghai and five in Guangdong.
Another factor that is equally important in building China’s robust AI ecosystem — which seems to be of more value — is the Tsinghua University which is considered to be the main incubator of China’s AI unicorns. The earlier mentioned paper notes that 26 percent of the founding teams and partners are graduates from this university.
Last May, the MOST New Generation AI Development Research Centre and the Chinese Academy of Science and Technology for Development published the China New Generation Artificial Intelligence Development Report which found the country to have 3,341 AI companies, second in number to the US. The report noted that China, the US and the UK are the three strongest contenders in AI innovation.
How China is capitalising on AI
China has taken its efforts a major step further. Now it has identified 17 chief areas for AI development. These include smart vehicles, intelligent service robots, intelligent drones, neural network chips and intelligent manufacturing among others. In November 2017, the China’s Ministry of Science and Technology chose Baidu, Alibaba, Tencent and iFlytek to take the lead in AI innovation development for self-driving cars, smart cities, computer optics for medical diagnosis and voice intelligence.
Essentially, this shows the country’s readiness to deploy AI technology applications in all walks of life from healthcare to manufacturing in a clear attempt to outpace efforts made by the US in the recent years. In line with this, AI unicorns such as Megvii moved into warehouse and logistics, Yitu focuses on medical imaging and document analysis, SenseTime developed in autonomous driving and iFlytek conducted demos for legal document analysis.
In particular, the work of Megvii reflects China’s AI capabilities in great length. Experts genuinely believe that it has impressive AI expertise after developing core algorithms and software. Interestingly, its technology is being used in Japan, South Korea, Southeast Asia, Europe, and the US prior to the company being blacklisted. The unicorn is now experimenting with warehouse and manufacturing AI technology.
Take healthcare in China. AI holds a prominent position in enabling automated diagnosis, medical imaging, robotic surgery and treatment—even if it is still in the early stages of development. The huge amounts of data that AI doctors will be potentially extracting could be used to help the field evolve in the future.
In fact, AI medical applications in China are predicted to grow at more than 40 percent and reach $2.5 billion by 2024. Industry experts forecast AI medical imaging to be the first market to commercialised in AI healthcare in China, especially because misdiagnosis and missed diagnosis are high in number, and the speed of manual diagnosis is limited. On a closer look, this ties back into a part of the country’s ambition to become the leader in certain aspects of AI technology by 2025.
The rise of AI unicorns
China’s big plans to have globally leading AI companies have birthed more than a dozen unicorns and private companies valued at over $1 billion, according to the research company CB Insights. These include SenseTime valued at $7.5 billion, Yitu and CloudWalk valued over $2 billion and iFlytek with a market capitalisation of $10 billion on the Shenzhen stock exchange.
The rise of these unicorns is a sign that the situation is transformative. Of course, the transformation in part can also be attributed to the Chinese government’s more than $1 billion investment in domestic startups between 2016 and 2018, with a deep focus on healthcare and AI, according to a report published by Oxford University in March 2018.
Also, China’s Ministry of Science and Technology provided funding of over $430 million for nearly eight AI-related research projects last year over a span of six months. Broadly speaking, these projects cover big data, high performance computing and human organs on chips.
The US hinders China’s AI developments
Although China’s AI companies built early success, their greatest roadblocks come from the US. Ideally, the US should not try to stop China from building on its AI potential. The first challenge is that the US has blacklisted some of the most critical Chinese AI companies like Megvii because its technology was allegedly being used by the Xinjiang police in a surveillance app. For security reasons, the Department of Homeland Security has warned American businesses to be careful when using drones made in China. This is hard hitting for Chinese AI companies as the repercussions to follow will curb their influence globally.
Second, a report published by the analyst companies IDC and Qbitai found that 60 percent of executives in China pointed to poor-quality data and scarce AI talent as reasons for difficulty in deploying AI.
Third, the big AI players in the US like Google, Facebook, Amazon, and Microsoft have myriad businesses to finance and support their AI efforts — quite opposite to the scenario in China. Alibaba, Tencent and Baidu, on the other hand, are deeply invested in AI which has now become a market heavily flooded with companies focused on AI as a business itself in China.
Will China still win the global race in AI?
Although both world superpowers are dominating global research and development in AI — the former lags the US in core AI areas, including hardware, research and algorithm, and industry commercialisation — with an exception to big data, observed Oxford University report.
Interestingly, Oxford’s Future of Humanity Institute published a report titled Deciphering China’s AI Dream which noted that extracting data is no problem for China where more than a billion smartphones are in use and had 840 million internet users last year. To second that, Dong Tao, vice chairman for Greater China at Credit Suisse Private Banking Asia Pacific, said in a report two years ago, that China will become the leader in AI because it lacks serious laws related to data protection. Its technology giants collect large amounts of data and the concept of sharing data between government agencies and companies is common — unlike the case in the US.
The Chinese are tech-savvy and the access to data will stimulate new ideas in the field of AI. In fact, China’s top search engine has pointed out that government support and population size is key to the global AI race. In this context, China has easy access to data despite being second in the world for hardware, research and algorithms — and the Chinese government’s efforts in pushing AI applications forward is highly competitive.
But will that be enough to support China’s AI ambition? The report found that data alone is not powerful enough for the country to achieve its target. A more important factor which is a lack of experienced AI researchers is anticipated to slow its efforts in AI innovation. And its biggest disadvantage is complexities in production of chips and microprocessors which might hold the country back from winning the global AI race.