CC BY
3072
УДК 327
LEGAL REGULATION OF ARTIFICIAL INTELLIGENCE IN THE MILITARY AND INDUSTRIAL COMPLEX (ON THE EXAMPLE OF CHINA AND THE EU)
© Abramovich Maria Mikhailovna'
'student, Siberian Federal university, Krasnoyarsk, Russia
Scientific advisor: Dzis Yulia Ivanovna
Candidate of Science (Political Science), Assistant Professor, Siberian Federal university, Krasnoyarsk, Russia
Abstract: The article is aimed at comparing the Asian and European approach represented by the EU and China to the development and formation of a regulatory framework in the field of artificial intelligence. At the beginning of the article the achievements in the field of general and military AI are presented. The analysis of the AI development concepts is carried out and the moral and ethical foundations that guide the participants in international relations are determined. In conclusion, the analysis of the prospects for the development of relations in this area is given.
Keywords: artificial intelligence (AI), legal regulation of AI, European Union (EU), China.
Developments in the field of artificial intelligence (AI) are considered almost the key direction of scientific and technological progress in the 21st century. But which countries and companies are leading on our planet in creating "thinking" and self-learning machines and computer programs? It is widely believed that North America is still ahead, but Asia is about to overtake it, while Europe is increasingly lagging behind.
Such a picture simplifies the situation too much, the balance of forces in this very dynamically developing sphere is much more complex. This is evidenced by two studies
published on January 31. One was presented by the Institute of German Economics (IW) in Cologne, the other by the World Intellectual Property Organization (WIPO) in Geneva, which is one of the specialized agencies of the United Nations.
The leading state in the field of artificial intelligence is China. This state is quickly ahead of all competitors in terms of innovative technologies and legal regulation. According to the US Department of Defense, China has decided to develop methods for introducing AI into future weapons systems. The Academy of Military Sciences of China was assigned to implement this program by combining the efforts of the military and industrial complex and private companies [1].
According to the level of development of artificial intelligence (AI) technologies, China is now in second place after the United States, and in ten years it intends to become the undisputed world leader in this field. The People's Republic of China has adopted a strategic state program for the development of AI until 2030. Its implementation is ensured by large-scale government funding, as well as funds from private technology companies closely associated with the Chinese state [2].
China's "New Generation Artificial Intelligence Development Program" sets three strategic goals for the country. First— by 2020, China's AI should be on par with similar industries in major developed countries. At the same time, the costs for the AI industry itself should amount to $ 22.5 billion, and for the development of related industries - more than $ 150 billion. The second goal is to achieve leading positions in some individual areas of AI by 2025, the fundamental industry will accumulate $ 60 billion, and related ones - $ 745 billion. Finally, by 2030, China will become the world's main center of innovation in the field of AI, by which time investments in the fundamental industry will amount to $ 150 billion, in the adjacent - $ 1.5 trillion.
In addition to the declared economic goals, the achievement of which should allow China to take a leading position in the global economic order of the future and compete with the United States and other developed countries from a position of strength, the government's "Artificial Intelligence Development Program" is designed to ensure the internal and external security of the country.
73
First of all, we are talking about creating a new system in China based on AI technologies to maintain socio-political stability in the country ("social trust system" or "social credit"). The plan for its formation is set out in the "Program for creating a social credit system for 2014-2020" adopted by the State Council in 2014. According to this program, a universal system of evaluation of citizens, companies and NGOs should be formed throughout the PRC. Each individual and legal entity will be assigned a "social trust" rating based on data on their social behavior. With the help of AI and in accordance with algorithms approved by the state, all available data about a person or company, including their online behavior, will be analyzed. Holders of a high rating will receive social and economic benefits; holders of a low rating will be subject to restrictions. Thus, the party expects to receive a tool that allows not only to exercise social control, but also to regulate social behavior [3].
But for the Chinese leadership, AI serves not only as a means of strengthening internal stability. It is also an important technology that will allow the People's Liberation Army of China (PLA) to conduct combat operations in the context of global informatization of military affairs. In January 2017, Chinese President Xi Jinping announced the creation of the Central Commission for Integrated Civil and Military Development. The task of the new body, led by the same Xi Jinping— is to integrate civilian technological solutions into the creation of new types of weapons for the PLA. And artificial intelligence is especially important here for the development of many systems. However, it is precisely progress in the civilian field of application that, according to Beijing, is of crucial importance for the development of the entire industry.
In the section "security measures" of the "New Generation Artificial Intelligence Development Programs", it is noted that along with the improvement of technologies, relevant laws and regulations should be developed, ethical standards for AI management should be established. The document recognizes that AI technologies will inevitably entail socio-ethical and legal problems. Among the specific products requiring legislative regulation, self-driving cars and autonomous robots are named.
In November 2017, the Chinese Academy of Information and Telecommunications, together with Tencent, released the book "Artificial Intelligence: a National Strategic Initiative". Two chapters are devoted to ethical issues and legal regulation of AI. The authors consider the problems of deviation of the actions of the machine from the behavioral patterns and goals pre-programmed in it by the developers. The problem of artificial superintelligence is being raised - when a machine acquires thinking similar to human thinking and can find goal-setting itself. At the same time, the authors note that it is not easy to solve these problems at the national level and international cooperation is necessary to develop measures regulating the basic moral and ethical principles of AI.
At the beginning of 2018, China published a White Paper on the Standardization of Artificial Intelligence. 30 organizations took part in its compilation: research institutes, analytical centers, private technology companies, including the Institute for Standardization of Electronic Technologies of the People's Republic of China, the Academy of Social Sciences of the People's Republic of China, as well as Baidu, Alibaba, Tencent and Huawei. The document describes the existing standards governing AI both in China and in other countries. The White Paper proposes a new comprehensive system of standards development that would regulate both the fundamental concepts of AI, and end products, and areas of technology application. A new structure is proposed for the development of AI standards in six categories [4,5]:
1) fundamentals (basic concepts, reference architecture);
2) platforms and support (smart sensors, chips, cloud computing, big data);
3) key technologies (face recognition, voice recognition, computer vision, augmented reality, etc.);
4) products and services (smart robots, autonomous drones, self-driving cars);
5) applications (smart manufacturing, smart logistics, smart cities, etc.). The sixth category of standards - safety and ethics - is end-to-end for all categories. It means that in parallel with the development of standards in the above five categories, standards should be developed
74
separately to ensure the safety and ethical use of these technologies. In the field of ethics, the document offers four basic principles:
• human interests,
• responsibility,
• transparency,
• consistency of rights and obligations. The latter principle is aimed, on the one
hand, at coordinating the commercial interests of companies and, on the other, at preserving the transparency of AI systems.
At the company level, China is also seeking to participate in the development of international standards in the field of AI. Thus, Baidu became the first Chinese company to join the "Partnership on AI" (PAI). This non-profit organization, which Amazon, Google, Facebook, IBM and Microsoft founded in 2016, develops ethical guidelines for AI research. The Artificial Intelligence Partnership unites academic, research, and public organizations; companies that develop and use AI technologies. The organization's task is to identify and study the best methods of using AI technologies, preventing possible violations of international conventions or human rights.
In turn, the EU aims to become a superpower in matters of ethics, morality and cybersecurity in connection with the development and use of AI. To achieve this goal, they have formed a strategy that includes:
1. Strengthening international cooperation in research and innovation, regulation of new markets, especially technological ones.
2. Development of a common, internationally recognized ethical and legal framework for the design, production, use and management of AI, namely robots and "autonomous" systems.
3. Development of new tools to solve the problem of foreign direct investment aimed at absorbing technology leaders with subsequent localization of acquired technologies.
4. Creating levers of political pressure for cooperation with China in the field of "smart" production, subject to significant progress in cybersecurity agreements.
5. Study of Chinese AI development goals, compatibility of government subsidies to companies working in this field with China's WTO obligations.
6. Development of measures to improve the efficiency of European companies in relation to intellectual production.
7. Collection of information on China's industrial policy by the European industry associations present in China and raising awareness of the EU governments on this issue to prevent China from developing independent national standards, which ultimately serve as barriers to trade and access to Chinese markets, and to promote international industry standards in China.
8. Find a balance between the benefits of ultra-high demand in China and the protection of basic ("core") knowledge and technologies.
At the same time, they face a number of problems that limit the development of the European Union in this area. For the first, AI technologies, big data and the Internet of Things pose serious problems for the concept of privacy in general and information privacy in particular; The European Union loses to the United States in the development of hardware and software, to China in ensuring effective control of access to data necessary for AI, and to the United Kingdom, according to its representatives, in developing ethics for the use of AI. The data of Europeans today actually belong to American IT giants; The Chinese industrial master plan "Made in China 2025", aimed at transforming the country into a manufacturing center of high-tech products, poses a threat to countries and companies from the EU; Cyberattacks of state or non-state origin carry increasing risks of a military nature, as well as manipulation of democratic public opinion and elections, economic espionage and technology theft; The information of the authorities is stored in different standards, which does not allow computer systems to use it fully.
The document presented in February 2020 by the European Commission is not the first to set out the strategy and vision for the development of AI in Europe. In 2018, the EU presented an AI development strategy and later a coordinated
75
action plan based on it, designed for a period up to 2027 [6].
The policies presented in February 2020 complement previous initiatives. The main emphasis is placed on the development of the AI ecosystem in the EU, the unification of the efforts of the EU member states and the development of the necessary legislative changes.
Feedback, suggestions were collected and public consultations were held on the new document, after which the European Commission invited the EU member states to revise the previously developed coordinated plan and adopt its new version by the end of 2020. Taking into account the proposals, ultimately the goal is to develop a European approach to AI [7].
The document focuses on two areas:
• Development of a set of policies and regulatory documents that would align efforts in the field of AI development at the European, national and local levels. They should allow businesses, together with the state, to mobilize their resources to create an "ecosystem of excellence", including a full production chain - from the start of research and development to the implementation of AI-based solutions, including in the segment of medium and small businesses (SMB).
• Development of key elements of the regulatory framework that would create a unique "ecosystem of trust". The regulatory framework in this case will be aimed at ensuring compliance with fundamental rights and consumer rights when using AI systems. First of all, we are talking about systems that carry high risks. The European Commission supports an approach focused primarily on people, the document emphasizes. That is why the highest executive
authority of the European Union in April 2021 proposed to limit the use of artificial intelligence (AI) in areas where technology can cause serious harm to society. Among the "high-risk" ways of using AI, the European Commission attributed, in particular, credit scoring, automated assessment of tests in the field of education, software for hiring employees, robotic surgery, as well as the use of algorithms in the fields of law and law
enforcement. To enter the market, such technological solutions will have to meet strict criteria of accuracy, security, data quality and openness to users and regulators. According to the proposed resolution, law enforcement agencies will not be able to use facial recognition technology in public spaces except in special cases (search for missing children, prosecution of criminals, prevention of terrorist attacks). The European Commission also proposed to completely ban the use of AI in "social rating" systems (as was mentioned earlier about the Chinese social credit system), as well as algorithms that "manipulate human behavior to circumvent his freedom of choice." EU experts suggest that AI solutions that pose a "minimal risk" to society should not be regulated in any way. According to them, "the vast majority of artificial intelligence systems" belong to this category.
The development of artificial intelligence generates competition and a battle between states for leadership in this area. However, whether these efforts are justified in real benefits for society or ignite new social conflicts. We clearly do not know what this cyber war will lead to and what difficulties humanity will face again, nevertheless, we clearly understand that the current problems facing each state collectively require a common solution.
References:
1. Brown D. Why China could win the new global arms race [Электронный ресурс]. URL: https://www.bbc.com/news/world-asia-china-59600475 (дата обращения: 25.01.2022).
2. Ковачич Л. Китайский опыт развития отрасли искусственного интеллекта: стратегический подход - Гонка технологий: перспективы искусственного интеллекта в России и Китае [Электронный ресурс]. URL: https://carnegieendowment.org/2020/07/0 7/ru-pub-82172 (дата обращения: 19.01.2022).
3. Ш-Ш^Ш^й^аШ^Ш^ШО
76
[Электронный ресурс]. URL: http ://www.gov.cn/zhengce/content/2014-
06/27/content 8913.htm
(дата
обращения: 25.01.2022).
4. Sheehan M. China's New AI Governance Initiatives Shouldn't Be Ignored [Электронный ресурс]. URL: https ://carnegieendowment.org/2022/01/0 4/china-s-new-ai-governance-initiatives-shouldn-t-be-ignored-pub-86127 (дата обращения: 25.01.2022).
5. Translation: Excerpts from China's 'White Paper on Artificial Intelligence Standardization' [Электронный ресурс]. URL:
http ://newamerica. org/cybersecurity-initiative/digichina/blog/translation-excerpts-chinas-white-paper-artificial-intelligence-standardization/ (дата
обращения: 25.01.2022).
6. A European approach to artificial intelligence | Shaping Europe's digital future [Электронный ресурс]. URL: https://digital-
strategy.ec.europa.eu/en/policies/europea n-approach-artificial-intelligence (дата обращения: 25.01.2022).
7. Mauritz K. EU Artificial Intelligence Act: The European Approach to AI [Электронный ресурс]. URL: https ://law.stanford. edu/publications/eu-artificial-intelligence-act-the-european-approach-to-ai/ (дата обращения: 25.01.2022).
В статье рассмотрен азиатский (Китай) и европейский (ЕС) подходы к развитию и формированию нормативно-правовой базы в сфере искусственного интеллекта. В начале статьи представлены достижения в области искусственного интеллекта в целом и военизированного ИИ в частности. Проведен анализ концепций развития ИИ и определены морально-этические основы, которыми руководствуются участники международных отношений. В заключении дан анализ перспектив развития отношений в этой сфере.
Ключевые слова: искусственный интеллект (ИИ), нормативно-правовое регулирование ИИ, Европейский Союз (ЕС), Китай.
Рукопись поступила: 14 апреля 2022 г. Submitted: 14 April 2022
ПРАВОВОЕ РЕГУЛИРОВАНИЕ ИСКУССТВЕННОГО ИНТЕЛЛЕКТА В ВОЕННО-ПРОМЫШЛЕННОМ КОМПЛЕКСЕ (НА ПРИМЕРЕ КИТАЯ И
ЕС)
Абрамович Мария Михайловна
студент, Сибирский федеральный университет, Красноярск, Россия
Научный руководитель: Дзись Юлия Ивановна
канд. полит. наук, доцент, Сибирский федеральный университет, Красноярск, Россия
