As an important driving force for the new round of scientific and technological revolution and industrial transformation, artificial intelligence ( ) is gradually penetrating and integrating into various fields of human society, having a significant and far-reaching impact on many aspects such as economic development, social progress, and international political pattern. However, in the process of broad and in-depth expansion of artificial intelligence applications, some risks and hidden dangers have been continuously exposed, such as privacy leakage, prejudice and discrimination, algorithm abuse, security issues, etc., which has aroused widespread attention from all walks of life. Especially the "human-computer alignment problem" encountered in development, that is, how to ensure that the artificial intelligence system is safe and controllable and conforms to human intentions and values.

In addition, the spread of inequality in the world of artificial intelligence will cause fairness and justice issues internationally, that is, artificial intelligence brings more conveniences to developed countries, and the negative impacts will affect developing countries. Therefore, the healthy development of international artificial intelligence is also an important factor in the stable development of the world. The differences in political systems and ideologies between countries cannot prevent consensus on AI governance.

Faced with the ethical risks in the development and application of artificial intelligence, countries and all sectors of society have begun to conduct ethical discussions, seeking paths and norms to deal with AI ethical risks to ensure the sustainable development of artificial intelligence. Therefore, artificial intelligence ethics (AI) has become a topic of concern to all sectors of society and has become a highly concerned research field.

Artificial intelligence governance policies have played a decisive role in the development of the artificial intelligence industry, the protection of public interests, future sustainable development, and international development trends. The policy chapter of the "2023 Global Artificial Intelligence Ethical Governance Report" aims to reflect the policy trends, current policy requirements and future policy reflections by mining and analyzing the policies of key countries. It clearly and specifically demonstrates the development history, focus and future trends of my country and overseas artificial intelligence ethical governance policies for organizers, participants and artificial intelligence users of the artificial intelligence industry and related industries, and plays a role in how it develops healthily and adapts to trend development.

A Yale University professor compared the different paths of artificial intelligence ethical governance in China, the United States and the European Union in 2022 and explained that the "Personal Information Protection Law" issued by China in 2021 is very forward-looking; the EU hopes to guide specific ethical practices through legal behavior, especially putting very high requirements for data protection. Many entities will be required to declare compliance with these regulations before entering the EU; the United States has formulated an algorithm responsibility law, hoping that companies will assume corresponding responsibilities, especially comply with the cyberspace management regulations, but are currently in a relatively stagnant state. Americans' obsession with innovation has forced many Americans to think about how to improve efficiency and take value factors into account to achieve a balance between productivity and ethical governance.

By 2023, the artificial intelligence ethical governance policies of various countries have been more comprehensive, systematic and precise. For example, the EU has issued the Artificial Intelligence Act, which first considers the core demands of regulatory "black box" technology, and breaks through the integration of artificial intelligence with the values of EU society. Secondly, while providing more detailed guidance, we create strict risk levels and provide a matching regulatory level for high-risk artificial intelligence technologies. The EU's Artificial Intelligence Act resolution marks an important step in the regulation of artificial intelligence and establishes new regulatory standards worldwide. The "Blueprint of the Artificial Intelligence Bill of Rights" issued by the United States has a more detailed explanation of human rights, and focuses on the integration and development of artificial intelligence with other industries, and has specific regulations and protections for different industries and jobs, and even covers the real estate market.

Unlike the EU, the US policy focuses more on serving the future development of the artificial intelligence industry and related linkage industries, and has established the leading position of American artificial intelligence in the world.

In 2023, my country also issued the "Interim Measures for the Management of Generative Artificial Intelligence Services", which is my country's first management method for generative artificial intelligence services. It not only provides specific regulations for the industry, but also different from Europe and the United States. While supervision, my country provides a broader space for development, which plays an important guiding role in the development of the industry.

As a technological revolution, artificial intelligence, from the perspective of technological empowerment, technological empowerment means that there will be some people who will be disabled, and there will be some people who will be lost in technological empowerment. Therefore, the Policy Section will be sorted out from the perspective of rights into three levels: human rights, property rights and sovereignty. For example, facial recognition issues at the human rights level, most of the identified persons in the identification process do not have the option to agree or not, which is controversial, and this is a human rights issue. In addition, the most obvious property rights issue is that the origin, source and property rights of generative artificial intelligence are unclear and the structure is very complex. A typical example of sovereignty issues is that Didi's listing was hindered by national data security issues. These will pose challenges to the future development of artificial intelligence.

In addition, in order to divide the human rights of the country into more detailed divisions, human rights are divided into privacy rights, fairness rights, and personal rights, and on this basis, further divided into operators and official institutions. In addition, in order to fully display policy trends and specific details, this article lists policies according to time during the process of policy splitting and classification. In addition, for the convenience of displaying synchronously with details, this article will provide a general description before the policy and list the original regulations below.

The following are some of the original texts:

If you want to read the full text of the report, please obtain it through the following figure: