🔬 Research summary by Dr. Marianna Ganapini (@MariannaBergama), our Faculty Director.
[Original paper by Huw Roberts, Josh Cowls, Jessica Morley, Mariarosaria Taddeo, Vincent Wang & Luciano Floridi]
Overview: This paper explores current China’s current AI policies, their future plans, and ethical standards they’re working on. The authors zoom in on China’s country-wide strategic effort, i.e. the ‘New Generation Artificial Intelligence Development Plan’ (AIDP). The strategic aims of the plan can be divided up into 3 main goals: international competition, economic development, and social governance.
It is no news to anyone that China is a leading force in AI, but what is their strategy in becoming an AI-superpower and what are they trying to achieve exactly? In this paper, the authors shed some light on the current Chinese policies and on China’s future plans for AI while also looking at some AI ethical standards that China is trying to develop. The bottom line is that China aims at becoming the AI world leader by rapidly developing breakthrough technologies that will completely revolutionize this field. However, China is poised to face several ethical challenges given its current authoritarian political system which sees technology also as a means to maintain control over its own citizens.
China and AI
The authors zoom in on China’s country-wide strategic effort, i.e. the ‘New Generation Artificial Intelligence Development Plan’ (AIDP) while focusing in particular on three “strategic areas”: international competition, economic growth, and social governance.
The AIDP was released by China’s State Council (their top administrative body) in 2017 and was the first “unified document that outlines China’s AI policy objectives” with the goal of making China “the world centre of AI innovation by 2030” and making AI the driving factor to propel China’s future economic and industrial developments.
The point of this plan is to function as a catalyst for busting the development of AI tech by private companies and local governments. The plan selects internationally established private companies as ‘AI National Champions’ (e.g. Alibaba, Baidu). “Being endorsed as a national champion involves a deal whereby private companies agree to focus on the government’s strategic aims. In return, these companies receive preferential contract bidding, easier access to finance, and sometimes market share protection”.
Similarly, local governments are empowered to incentivize the development of new technologies while also trying to fulfill “national government policy aims”. It is important to stress that private companies and local governments are given a lot of leeway in how to proceed: they are only provided “few specific guidelines”. And “[t]his enables companies to cherry-pick the technologies they want to develop and provides local governments with a choice of private-sector partners”.
The strategic aims of the plan can be divided up into 3 main goals: international competition, economic development, and social governance.
China’s main focus in international competition is to develop breakthrough AI military technologies to overtake the US. For instance, they have been developing cyber warfare and cyberattack strategies to gain valuable knowledge and intelligence and they clearly see technology and AI in particular as a way to gain strategic military advantages over the US. At the same time, however, top officials in China seem to be aware of the dangers of AI in fostering a “potential military escalation” and see the need for cooperation in mitigating potential risks (especially the risks posed by autonomous lethal weapons).
AI is considered as the future driving force behind China’s economic growth which is key to keep up with China’s economic and industrial expansion of the last few decades. The potential of AI as an economic force comes also with some added risks for the economy as it can disrupt the labour market and negatively affect low- and medium-skilled jobs. Though China is preparing for these structural changes, “[e]stimates show that, by 2030, automation in manufacturing might have displaced a fifth of all jobs in the sector […]. These changes are already underway, with robots having replaced up to 40% of workers in several companies.” This process may actually worsen China’s domestic economic inequalities.
The AIDP explicitly tackles the social challenges China faces, from economic inequality to the lack of a “well-established welfare system” and a rapidly expanding “environmental degradation”. Technology is seen as key in producing meaningful changes in the healthcare sector, and AI is perceived as a tool to address some of China’s environmental problems — especially those related to pollution in the air.
Similarly, AI will be used to administer justice in a potentially more efficient and transparent way with the stated goal of fixing some of the long-term problems of China’s judicial system. This attempt has already raised some eyebrows: using AI to administer justice has often led to even more injustice and unfairness. China is using technology to structure more efficiently its Social Credit System, a system that needs an extensive amount of personal data.
Social governance in China also means smart cities and surveillance technologies. Possibly, the most egregious example of this is the surveillance program adopted in the autonomous region of Xinjiang where so-called “potential terrorists” were tracked through facial recognition and other invasive surveillance technologies.
Ethics of AI in China
Given the risks and opportunities of massive development of AI technologies, “the AIDP outlines a specific desire for China to become a world leader in defining ethical norms and standards for AI”/ Based on this, ethical principles and guidelines were put forward which “bear some similarity to those supported in the Global North” in their emphasis on transparency, privacy, accountability and respect for human welfare. Yet there will be some important differences in how these principles are understood and applied because, as the authors point out, “China’s AI ethics needs to be understood in terms of the country’s culture, ideology, and public opinion”. For instance, after its past timid efforts to protect data and personal privacy, China is now trying to enforce some privacy regulations.
However, China is now also struggling to define exactly what type of data needs to be protected and how. The main ethical challenge, in this case, is how to square the idea of personal privacy within an authoritarian political system in which the government is de facto not constrained by regulations meant to protect its citizens (as seen in the mass surveillance program in use).
Similarly for medical ethics, China’s main goal is “societal welfare” rather than individual wellbeing. According to the authors, that boils down to the idea that personal medical data will be shared widely and used by the medical community to find cures that will benefit society at large with little concern for the privacy and the rights of single individuals.
In conclusion, though there seems to be a growing concern for the ethical challenges posed by AI, China is still very much struggling to tackle these problems given its current political and social system which often uses morally dubious strategies and tools to maintain control over its own citizens.
Between the lines
We believe this paper represents an important step in the right direction of developing a well-informed analysis of the risks and opportunities of AI in non-Western countries. It is important to acknowledge the ethical limitations of the development of AI in China while also trying to objectively report the current efforts being made to overcome these limitations. What is now needed is a careful comparative analysis of the shortcomings of the Global North compared to China because it is unclear that the problems we see in China are not present in some form also in western countries. For instance, it would be important to understand how Europe and the US fare compared to China on issues such as privacy and transparency in AI.