Artificial Intelligence in China

From SI410
Jump to: navigation, search

In the 21st century, Artificial Intelligence(AI) has been a vital part of China's technology industry. It's being used across myriads of sectors including education, video games, military, automotive, healthcare, government, manufacturing, finance, cybersecurity, and art, among many more. It's implemented in many applications including chatbots(such as Xiaoice), search engines(such as Baidu), drones, self-driving cars, and China's Social Credit System. However, China wasn't always this receptive to new technology. In the 1950s, the Chinese government saw the advancement of science and technology as a negative influence on society and its political campaigns, rejecting any related research projects. It wasn't until 1978 when the first National Science Conference was held in Beijing, China and the leader of the country, Deng Xiaoping pushed the advancement of science and technology to modernize the economy. Since then, many citizens were encouraged to work on AI projects.

In 2017, China's State Council issued a centralized national strategy to promote the development of AI, envisioning a future where it will aid in industrial transformation, better social governance, and maintaining social stability. [1] China plans to incorporate the technology in almost all industries to become a global superpower for AI by 2030. However, because AI has been advancing so rapidly, people consider it to be a danger to humanity if it progresses without being subjected to rules and regulations. One example include China's mass surveillance systems - AI has been a prevalent part of China's citizens day to day life with millions of facial recognition cameras installed throughout the country and phone tracking capabilities. This has raised many questions of privacy and the devaluation of humanity, domestically and globally, which increasingly pushed for the need for governance principles over the development and deployment of AI. In 2017, the government created an ethics board and the first official Governance Principles for the New Generation of Artificial Intelligence to ensure the responsible development of the technology. In 2019, many universities and companies came together to release the Beijing AI Principles. Many ethical issues still arise with AI and its governance in China today but it continues to help solve many challenging problems in society, influencing billions of lives.

Early History of Science and Technology Policy in China


In 1950s China, both the Chinese government and its scientists prevented the advancement of cybernetics, which according to Kolmogorov was science concerned with the study of systems of any nature which are capable of receiving, storing and processing information. Cybernetics was deemed a bourgeois pseudoscience and many scientists rejected its influence. Moreover, because China was faced with enormous damage from WWII at the time, it did not have the resources to focus on science and technology. In the 1960s, political campaigns such as the Great Leap Forward and the Cultural Revolution led to steep economics declines, exiles of many intellectuals, and paralyzed education systems, which further led to the decline of technology and science advancement. The "Gang of Four", a powerful political faction in China, criticized scientific theories such as Einstein's theory of relativity, Weiner's cybernetics, and the Big Bang Theory, preventing people from indulging in scientific and technological research. [2]

Turning Point

However in the 1970s, after the Gang of Four lost its power and the Cultural Revolution ended, the year 1978 became a turning point for the development of science and technology in China. The People's Republic of China held the first National Science Conference that year in Beijing to encourage the development of scientific and technological research. Deng Xiaoping, the leader of the People's Republic of China at the time, stressed in his opening remarks that science and technology were productive forces, and the key to China's modernizations was the modernization of science and technology.[3] The first cybernetics and AI research project was proposed here as a result.

Recent History

Quick recent history of AI in China

Government Supports AI

China is catching up fast in the AI industry and one of its biggest reasons is its collaboration between the Chinese government, private sector, investors, and academia. With the government funding the industry through developing national AI research centers, adding university courses on AI, enforcing tax policies benefitting technology companies developing or adopting AI, and initiatives to attract more AI talent and projects, the government's top-down approach help propel China's goal to be an AI superpower. Moreover, the country's millions of online connected users and surveillance on its citizens establish the biggest consumer database that any institution can use to train its AI technologies. [4]

Research Papers

China is continuously investing in AI research domestically and globally - it has published the highest amount of AI research papers so far, but numbers show they’re cited significantly less than US and EU papers. As a result, China has invested in improving the quality of its AI research papers by pushing three initiatives: [5]

1) China’s New Generation AI Development Plan, which calls for China to have made significant breakthroughs in AI theory by 2025

2) The Chinese government has created research centers, including the Artificial Intelligence Research Center, which has more than 100 employees.

3) The Chinese Ministry of Industry and Information Technology plans to allocate $950 million RMB annually to fund strategic AI projects.

New Generation AI Development Plan

On July 20, 2017, the state council released an official Plan for the Development of New Generation Artificial Intelligence - the first of its kind to address AI as a national strategy in China. It outlined goals to build a domestic AI industry worth 150 billion RMB in the next few years and become the leading AI power by 2030. [6]

By 2020, the government expects “its companies and research facilities to be at the same level as leading AI countries like the United States” and the technology to become an important driver for economic growth [6]

By 2025, there will be major breakthroughs for fundamental AI theories and China will become the world’s leading level for some sets of AI technologies. AI will be applied in most areas such as smart manufacturing, smart healthcare, smart cities, smart agriculture, and national defense [6]

By 2030, China will become the world’s leader and global center for all AI theories, technologies, and applications. AI will also be deeply integrated into most industries [6]

As a result of this plan, more than 1000 AI companies in China pushed out its first commercial applications a year later and the State Council allocated 2.1 billion RMB on an AI industrial park in Mentougou district.

AI Applications in China

AI traffic cameras tracking car plates and people crossing

China has world-leading companies in computer vision, speech recognition and natural language processing, including SenseTime, Unisound, iFLYTEK, and Megvii, says Zheng Nanning, director of the Institute of Artificial Intelligence and Robotics at Xi’an Jiaotong University. [7] But this is only the tip of the iceberg when it comes to how AI is applied and utilized in China.

Crime and Traffic Safety

The largest algorithm provider in China as well as the fifth-largest AI platform, SenseTime developed a facial recognition system called DeepID that lets China's police departments analyze video footage to catch criminals, track car plates to support traffic management, recognize traffic incidents to alert authorities, and track driver's motions to identify if they are falling asleep behind the wheel among many other applications. [8]

Social Credit System

The Social Credit System is a reputation system being used in China to assess each of its citizen's trustworthiness. It uses facial recognition technology, big data analysis, and AI to track its citizens and assign a score to each of them. People's social credit scores get lowered if they are dishonest, commit fraudulent financial behavior, play loud music, violate traffic rules, fail to correctly sort personal waste[9], fraudulently use other people's ID cards, or steal among many other behaviors that are listed as negative factors in the credit ratings. Positive factors on the other hand increase credit scores and those behaviors include donating to charity, donating blood, volunteering for community services, and so on. [10] If people have a low social credit score, the government can bar them from traveling, getting a mortgage, landing a job, and shopping online. [10]


Students in China using AI learning on their computers

Personalized Learning

Over 1 billion RMB was invested in AI education in 2019 and one of these companies was Squirrel AI Learning. It uses adaptive learning, an education technology that utilizes AI to understand what learning materials a student is struggling with, then respond in real-time providing them with online support, such as more learning materials on the topic, and face to face interaction until they master the subject.[11] Currently, the company has opened over 1700 schools in more than 200 cities across China using this technology.

Facial Recognition in the Classroom

AI, brain wave trackers, and cameras are being installed in classrooms to enhance teaching methods and monitor students' attention levels. By using facial recognition technology, brain wave trackers, and a live platform for students to point out what they understand and don't understand in class, schools could design curriculum around the students' attention patterns, better assess attendance, improve student engagement and learning activities, proctor exams, and enable educators to measure the effectiveness of their teaching methods. Parents could also get live information about their children's learning patterns, attention levels, and engagement in class. [12]


In 2018, Chinese financial institutions invested 160.4 billion RMB in technology, an increase of 10 percent over 2017, of which AI hardware and software-related investment accounted for 10.4 percent. The banking industry accounted for 70 percent of all market purchases of AI, becoming the biggest investor in AI-related applications. [13] Because banks have massive amounts of data on people's financial transactions, market analysis, customer profiles, and risk control, they're able to provide an environment for AI to flourish - such as in precision marketing, intelligent risk control platforms, facial recognition cameras, and document identification machines. [13]

Baidu testing their self-driving cars in Beijing

Autonomous Vehicles

When China’s government declared in 2017 that being the global leader in autonomous technology was a national priority, technologies like AI, 5G wireless networks, robotics, and smart cities became a priority associated with it as well. [14] Since then, many companies have worked to create autonomous vehicles such as buses, cars, robo-taxis, and drones using AI.

Self-Driving Cars and Robotaxis

Among dozens of autonomous driving companies, Baidu, one of the biggest internet and AI companies in China, became the first to secure licenses in July 2019 to test self-driving cars in Beijing. It was the first city in China to allow autonomous road tests and by the end of December 2019, the city has allowed over 150 roads totaling more than 500 kilometers, or 310 miles, in length to be used for these tests. [15]

Controversies and Ethical Issues

Mass Surveillance

To track everyone in the country, the Chinese government is using facial recognition cameras installed on streets and inside buildings to help police grab the identities of people wherever they travel, find out who they are meeting with, and identify who does and doesn’t belong to the Communist Party. By 2018, they had installed close to 200 million surveillance cameras across the country, which amounts to approximately one camera per seven citizens. [16] China has become the world’s biggest market for security and surveillance technology, with analysts estimating the country will have almost 300 million cameras installed by 2020. [16] On top of that, the Chinese government has adopted surveillance drones, robot police, internet surveillance, and big data collection targeting online social media platforms to monitor its citizens, which has given the police and government authorities immense power to control people at the expense of personal privacy. [17]

Ethnic minorities inside a Xinjiang re-education camp

Monitoring Ethnic Minorities in Xinjiang

In Xinjiang, its residents especially those from the Ughyer ethnic group are strictly monitored by surveillance apps and facial recognition cameras, ensuring they do not practice certain religious acts. The identities collected by the cameras are matched with individual profiles that include data about their biometrics, such as DNA samples and voice samples, and the people are algorithmically rated on a level of trustworthiness based on their profiles. [18] This brings up many issues as people could not opt-out of being surveilled, losing their privacy and ability to express their cultural identity. Many were arrested as a result of practicing their religion.

Predictive Policing and Re-education Camps

Since 2017, the government has set up re-education camps and an AI system that the government claims can predict individuals prone to "terrorism" and detect those in need of re-education [19], forcing ethnic minorities in China such as Uyghurs, Kazakhs, and Kyrgyz to comply to pre-existing biases from the government, technical biases from the algorithm, and emergent biases from the geographical region and ethnicity the technology is used on. People were tortured and forced to undergo psychological indoctrination in the camps - like studying Communist propaganda, learning Mandarin Chinese, as well as praising the Chinese government - causing many human rights violations. [20]

Students with brain-wave trackers and cameras watching them in the classroom

Privacy in Education

With facial recognition data about each student in the classroom, students and their families are worried about what data points are being shared with their future schools and universities. Specifically, what if data about how focused each student was in their class gets shared with future schools and universities, damaging their chances of getting their dream education? [21] This raises questions on how privacy has been an ongoing issue with AI as not all companies are transparent with what data they're collecting and sharing and what's not.

Face-Swapping in Online Payments

Zao, a face-swapping technology, first intended for users to swap their faces with friends and family for fun later posed a security threat to many of China's facial recognition payment systems as users could pose as other users to use their bank cards. [22] Many payments in China are cashless, relying on mobile apps, facial recognition cameras, and online payment systems, which means information security and cybersecurity are increasingly important behind the advancement of these technologies.

Military Drones

Since 2019, China's military contractors have started to sell autonomous killer drones to customers in the Middle East. One company, called Ziyan is marketing an autonomous helicopter drone armed with a machine gun to the governments of Pakistan and Saudi Arabia, signaling a future of an AI arms race, despite not aligning with values from China's ethics board and AI governance principles on doing good for humanity. [23]

AI Governance Principles in China

Governance Principles for the New Generation of Artificial Intelligence

Many AI technologies and applications were developed as a result of the national strategy but an ethics board to govern the rapid development was still not created yet. On June 17, 2019, China’s Ministry of Science and Technology published the first official Governance Principles for the New Generation of Artificial Intelligence after several rounds of discussions with the Ministry’s New AI Governance Expert Committee. They proposed eight principles to ensure the development of responsible AI: [24]

1) Harmony and friendship

AI must align with society’s values and ethics; misuses and abuses must be prevented and prohibited. [24]

2) Fairness and justice

AI must protect the interest of all stakeholders and promote equal opportunities; prejudice and bias must be eliminated in data acquisition, algorithm design, product research, technology development, and the application itself. [24]

3) Inclusiveness and sharing

AI must aim to enhance the well-being of humanity; AI education must be strengthened and accessible to disadvantaged people and industries; data and platform monopolies must be prevented and open cooperation is encouraged. [24]

4) Respect for Privacy

Personal privacy should be respected and protected; privacy standards should be established for the collection, storage, processing, and use of personal information in all aspects of product development. [24]

5) Security and Controllability

The public should have the right to know what data is being shared and the right to choose what to share; AI development should be transparent and ways to supervise, manage, track, and monitor those systems should be developed. [24]

6) Shared Responsibility

AI developers, users, and all stakeholders have the shared responsibility to respect laws, ethics, and norms; an accountability system will be developed to assess the role of each. [24]

7) Open Collaboration

Interdisciplinary and cross border collaboration is encouraged and interactions in the development and governance of AI are promoted among the general public, research institutions, educational institutions, social organizations, enterprises, government departments, and international organizations. [24]

8) Agile Governance

Management and governance systems will be constantly updated and improved throughout the AI product and services life cycle; research of potential risks related to the AI will be done to ensure it will continuously enhance the well-being of humanity. [24]

Beijing AI Principles

In May 2019, the Beijing Academy of AI (BAAI), Peking University, Tsinghua University, Institute of Automation and Institute of Computing Technology in Chinese Academy of Sciences, and an AI industrial league involving firms like Baidu, Alibaba and Tencent, released the Beijing AI Principles calling for the “the construction of a human community with a shared future, and the realization of beneficial AI for humankind and nature.” [25]

The Principles are separated into three sections:

1) Research and Development

AI should be designed and developed to do good, serve humanity, be responsible, be secure and be able to control risks, be ethical, diverse and inclusive, as well as be an open platform for sharing and promoting equal opportunities. [25]

2) Use

AI should be used wisely and properly such that the users have the knowledge to operate the systems according to its design, avoiding misuse and abuse. Users should be informed about the technology's implications and be able to consent to AI's impact on their rights and interests. Stakeholders should be trained to adapt to the implications of AI psychologically, emotionally, and technically. [25]

3) Governance

AI should optimize employment, promote harmony and cooperation domestically and internationally to avoid a malicious AI race, and be adaptable to new rules and regulations. AI principles, policies, and regulations should actively adjust to the development of AI and continuous research on the potential risks of AI should be encouraged. [25]

Overall, these principles focus on benefitting all of humanity and the environment; serving human values such as privacy, dignity, freedom, autonomy, and rights; continuous focus on AI safety and security; inclusivity; openness; supporting international cooperation and avoiding a “malicious AI race”; and long-term planning for more advanced AI systems, among others. [25]


  1. A Next Generation Artificial Intelligence Development Plan,
  2. Jieshu. “The Early History of Artificial Intelligence in China (1950s – 1980s).” Jieshu's Blog, 7 Nov. 2018,
  3. 张若琼 . “March 18-31,1978: The National Science Conference Is Held in Beijing.” China,
  4. Faggella, Daniel. “AI in China – Recent History, Strengths and Weaknesses of the Ecosystem.” Emerj, Emerj, 5 Apr. 2019,
  5. Daxue_Writing. “The History of Artificial Intelligence (AI) in China: Daxue Consuliting.” Daxue Consulting – Market Research China, 17 Jan. 2020,
  6. 6.0 6.1 6.2 6.3 “State Council's Plan for the Development of New Generation Artificial Intelligence.” China Innovation Funding,
  7. O’Meara, Sarah. “Will China Lead the World in AI by 2030?” Nature News, Nature Publishing Group, 21 Aug. 2019,
  8. Marr, Bernard. “Meet The World's Most Valuable AI Startup: China's SenseTime.” Forbes, Forbes Magazine, 17 June 2019,
  9. Huang, Echo. “Garbage-Sorting Violators in China Now Risk Being Punished with a Junk Credit Rating.” Quartz, Quartz, 10 Jan. 2018,
  10. 10.0 10.1 “How China's 'Social Credit' System Blocked Millions of People from Travelling | CBC Radio.” CBCnews, CBC/Radio Canada, 7 Mar. 2019,
  11. Hao, Karen. “China Has Started a Grand Experiment in AI Education. It Could Reshape How the World Learns.” MIT Technology Review, MIT Technology Review, 3 Aug. 2019,
  12. Chinese School Used AI to Monitor Students - West,
  13. 13.0 13.1 Synced, and Machine Intelligence | Technology & Industry | Information & Analysis. “China's Financial Services Industry Is Banking on AI.” Synced, 4 Jan. 2020,
  14. Schenker, Jennifer L. “China's Drive To Dominate Autonomous Cars.” Medium, The Innovator News, 27 Aug. 2019,
  15. Wiggers, Kyle. “Baidu Secures Licenses to Test Self-Driving Cars in Beijing.” VentureBeat, VentureBeat, 1 Jan. 2020,
  16. 16.0 16.1 Mozur, Paul. “Inside China's Dystopian Dreams: A.I., Shame and Lots of Cameras.” The New York Times, The New York Times, 8 July 2018,
  17. Mozur, Paul, and Aaron Krolik. “A Surveillance Net Blankets China's Cities, Giving Police Vast Powers.” The New York Times, The New York Times, 18 Dec. 2019,
  18. author., Wang, Maya. "Eradicating ideological viruses": China's campaign of repression against Xinjiang's Muslims. OCLC 1052124054
  19. “How China's Government Is Using AI on Its Uighur Muslim Population.” PBS, Public Broadcasting Service,
  20. Campbell, Alexia Fernández. “China's Reeducation Camps for Muslims Are Beginning to Look like Concentration Camps.” Vox, Vox, 24 Oct. 2018,
  21. “AI Is Watching China's Students but How Well Can It Really See?” South China Morning Post, 15 Sept. 2019,
  22. Chengevelyn. “Growing Backlash in China against A.I. and Facial Recognition.” CNBC, CNBC, 17 Jan. 2020,
  23. Tucker, Patrick. “SecDef: China Is Exporting Killer Robots to the Mideast.” Defense One, 5 Nov. 2019,
  24. 24.0 24.1 24.2 24.3 24.4 24.5 24.6 24.7 24.8 “Publication of the ‘New Generation AI Governance Principles – Developing Responsible AI.’” China Innovation Funding,
  25. 25.0 25.1 25.2 25.3 25.4 Beijing AI Principles, “AI Policy - China.” Future of Life Institute,