Strengthening ethical governance in the field of artificial intelligence in China
The Central Committee of the Democratic Progressive Party (DPP) intends to submit to the 14th session of the National Committee of the Chinese People's Political Consultative Conference (CPPCC) a proposal on strengthening ethical governance in the field of artificial intelligence in China.
The Central Committee of the Democratic Progressive Party points out in the proposal that the report of the 20th National Congress of the C*** proposes to build a new growth engine in artificial intelligence and other fields. With the rapid development of AI technology, the ethical challenges facing science and technology are becoming more and more prominent, and the recent ChatGPT, which has exploded all over the Internet, has once again triggered the international community's concern about ethical issues in the field of AI. The Opinions on Strengthening Ethical Governance of Science and Technology issued by the Central Office and the State Office have made artificial intelligence one of the key areas where ethical governance of science and technology needs to be strengthened. Ethical governance in the field of artificial intelligence in China still faces many problems.
One is the relative lag in system construction.
In terms of system, there is a relative lack of laws and regulations that can be implemented on the ground, and the bottom line of compliance and disciplinary efforts are not clear enough. In terms of management, the lack of coordination is more prominent, the division of responsibilities is not clear enough, and there are still gaps in the review and supervision of ethical governance.
The second is that capacity building is relatively weak.
China's ethical governance in the field of artificial intelligence is still in its infancy, and there are problems such as irrational composition of personnel, weak ethical awareness, insufficient professional capacity, and imperfect working rules and operation norms, and the capacity building needs to be further enhanced.
Thirdly, the participation of enterprises in governance is insufficient.
China's promotion of AI enterprises to participate in ethical governance on their own is still in a preliminary stage of exploration, and there is a lack of clear regulations and operational institutional safeguards for the boundaries and mechanisms of enterprise participation, as well as a lack of willingness to participate in governance and a lack of experience and capacity.
Solutions:
One is to build a governance system with multi-party participation.
The Ministry of Industry and Information Technology (MIIT) will take the lead in designing a top-level structure for ethical governance in the field of AI in terms of leading institutions, organizational structure, division of labor in the field, and social synergy, so as to make clear the obligations and responsibilities that need to be fulfilled by all parties in the AI application chain. Promote the construction of laws and regulations, make clear the bottom line of legal compliance, and clarify the disciplinary efforts for data abuse and infringement of personal privacy. At the same time, through national standards, industry standards, enterprise self-regulation rules and other "soft law" methods, to guide enterprises to comply with innovation. Adopting experts from AI enterprises as members of the National Science and Technology Ethics Committee's AI Sub-Committee, so as to better study and solve ethical problems faced by enterprises in technological innovation. Utilizing the science and technology ethics consulting role of the National Committee on the Governance of New-Generation Artificial Intelligence to conduct forward-looking research and judgment on ethical risks that may arise in the development of technology. Utilizing the role of national science and technology think tanks and related research forces, it will conduct systematic research on the ethical issues of artificial intelligence and provide decision-making advisory support.
The second is to carry out multi-dimensional and multi-level ethical construction.
Carrying out value-sensitive design for the whole life cycle, in the research and development stage, analyzing the ethical problems that may arise from the new technology, and controlling the potential ethical risks from the source; in the design and manufacturing stage, weighing the cost of the new product to be accepted by the society morally against the benefits it will bring; in the experimental and promotional stage, making adjustments to the new product according to the society's value choices, and improving the compatibility of the new product with the social In the stage of experimental promotion, new products are adjusted according to the value choices of society to improve the compatibility of new products with social values; in the stage of deployment and application, the degree of user recognition and acceptance of new products is analyzed, and the ethical risks triggered by the application process are examined and reflected. Carry out ethical regulation of multi-application scenarios. The ethical problems arising from AI in different application scenarios and the focus of the work to be governed are different, and their ethical regulation needs to be studied separately. For example, daily applications such as autonomous driving and medical diagnosis require enterprises and the public to carry out ethical regulation based on the characteristics of the application scenarios and actual ethical issues; the ethical risks brought about by the use of AI technology in military confrontation need to be restrained through international agreements. To carry out ethical regulation of cross-field, artificial intelligence and other disciplines after the cross-application of the formation of new fields, such as artificial intelligence and neurotechnology cross-brain-like organs, need to be combined with the characteristics of the cross-field to carry out differentiated ethical regulation of the study.
Thirdly, enterprises are encouraged to participate in governance autonomously.
To clarify the driving force for enterprises to carry out ethical governance, mainly including: to become an enterprise that abides by ethical rules and establishes a corporate image in the industry; the enterprise's products should not only be technologically competitive, but also ethically and morally trustworthy to consumers and the industry; and to introduce responsible new technologies globally to lead the direction of AI development, and so on. Cultivate the awareness of corporate ethics, strengthen the training and learning of AI ethics for corporate executives, and improve their ethical knowledge and judgment; for technical R&D personnel, advocate responsible technical R&D, and establish the ethical concept of "trustworthy and responsible". Drawing on the experience of Microsoft and other international AI companies in setting up ethics committees, we support the establishment of ethics committees within companies to examine the risks of uncontrollability, infringement, discrimination, and liability that may be brought about by AI, and to provide the necessary conditions for them to carry out their work independently.