Research on artificial intelligence data protection in China

Yeoh Roger

Abstract: Artificial intelligence (AI) has become the core force driving the development of global science and technology and economy recent years, and the importance of data as its core asset is self-evident, but it also brings about a series of data protection problems. Though China has constructed a number of data protection regulatory system, there still exists some lack of strategies, which makes China fall behind the international general management system a bit. The essay analyzes China's AI development strategy, data protection policies and regulations, and regulatory enforcement mechanisms, as well as analyzes existing problems, such as poor departmental coordination and insufficient international compliance. The research results show that China needs to promote reforms of artificial intelligence and improve the relevant systems in various aspects, such as data classification and hierarchical protection, cross-border data management and the like, in order to promote the healthy and sustainable development of the AI industry.

1 Introduction

Artificial intelligence technology has become the leading force for the development of global science and technology and economy. As a key resource for AI development, data protection is becoming the focus of international competition and social governance. In recent years, China has adopted strategic arrangements such as the Development Plan for a New Generation of Artificial Intelligence and the Three-year Action Plan for Promoting the Development of Artificial Intelligence Industry, which have established a complete system covering technology research and development, industrial coordination and talent training. Through legislative guidance, tax incentives, cooperation between industry, university and research institutes, and talent introduction, the government has promoted AI technology to leap from "synchronous catch-up" to "local leading". However, the rapid iteration of AI technology and the large-scale application of data also bring serious challenges: frequent domestic data leakage incidents (such as the Ctrip data breach case in 2016), loopholes in cross-border data flow supervision, and inefficient law enforcement due to the poor coordination mechanism of multiple departments. At the same time, international rules such as the EU's General Data Protection Regulation (GDPR) put higher requirements on Chinese companies' overseas compliance, while the domestic data management system is not sound, and it is urgent to seek a balance between technological innovation and privacy protection.

Focusing on the institutions and practices of AI data protection in China, this study mainly focuses on the following core issues: First, how to effectively balance the technological development and security requirements of the current AI data protection framework. What loopholes exist in the implementation of existing laws such as Network Security Law and Data Security Law? Second, what barriers multi-agency collaboration (e.g., CAC, MIIT, Ministry of Public Security) faces in AI data regulation. How to optimize cross-departmental data sharing and responsibility division. Finally, how to build a data protection system in line with China's national conditions while learning from the EU's GDPR experience. What room for improvement exists in cross-border data flow management and data subject rights protection?

Through systematic analysis, this paper proposes the following innovative perspectives and solutions: Firstly, in the strategic and regulatory level, the evolution of China's AI development strategy and data protection legislation is sorted out, and the transformation logic from "technology catch-up" to "norm leading" is revealed. Combined with international comparison, specific suggestions are put forward to improve the classification and classification protection of data and the security evaluation mechanism of cross-border flows. Secondly, aiming at the coordination problem of the regulatory mechanism, the barriers to information sharing and the ambiguity of rights and responsibilities in multi-department collaboration are analyzed, and the reform scheme of establishing a cross-department data sharing platform and clarifying the regulatory boundaries of emerging fields (such as intelligent driving) is proposed. The Ctrip data leak case is taken as a case to demonstrate the necessity of strengthening enterprise compliance and technology empowerment regulation. In addition, in terms of protecting the rights and interests of data subjects, we discuss the transparency of privacy policies and the strengthening of user data control rights, and combine the practices of domestic and foreign enterprises (such as Google's interactive privacy policy) to put forward the implementation path to improve public trust. Finally, from the four dimensions of legislation refinement, technology evaluation, international cooperation and public education, a "four-in-one" optimization framework for AI data protection in China is constructed, which provides reference for policy making and industry practice.

Through the combination of theoretical analysis and empirical research, this paper provides systematic solutions for the system improvement and technical innovation of AI data protection in China, and helps the healthy development of digital economy and social stability. This study emphasizes that China needs to take into account local practices in international competition, and build a safe and inclusive AI data governance system through legal improvement, technological innovation and social co-governance.

2 Artificial intelligence development strategy and data protection status in China

2.1 The strategy of the development of artificial intelligence in China

Since the dawn of 21st century, numerous emerging technologies have been taken out of imagination and arise rapidly than any of the former centuries, such as the innovation of smartphones, 3D painting as well as Gene editing technology CRISPR - Cas9. Besides, in the last few years artificial intelligence has been focused on development and improvement as Gao (2008) stated. It can be said that artificial intelligence plays the most crucial and leading role in present tendency of scientific and technological progress. Enormous amounts of technology giants delved into the research and competition in the artificial intelligence market, such as, Apple, Google and Meta.

As Zhang Huilian (2024) stated, not only the effort from companies would determine the market of artificial intelligence, governments also have a responsibility that can not be ignored. Governments have a role to play in strategic placement and resource integration, basic research and talent development, and market regulation and security, which the mission could not be taken by any individual companies. In addition, countries also enjoy the benefits from the arrangement of AI industry layout. Slow economic development can be digitally transformed with AI technology to bring new propulsion to economic development. Hence the evolution of artificial intelligence leads to the technology and innovation competition intensifies among different nations gradually.

Therefore, China has already made efforts to plan and design the strategy of the domestic development of artificial intelligence in order to keep pace with the times as Allen (2019) stated. Like many other countries, China takes hand into the strategy through a series of initiatives. First of all, the Chinese government enacted Three-Year Action Plan on Promoting the Development of Artificial Intelligence Industry (2018-2020) and the New Generation Artificial Intelligence Development Plan and a series of fundamental legislation recent years. By setting a set of milestones, China starts from technical synchronization to partial leading, and aims to become one of the vital centers of artificial intelligence innovation in the future finally. The direction of domestic industry development is anchored in the legislation initially. Secondly, the government has offered tax incentives in order to reduce the enterprise cost; Specialized departments have also set up special funds to support research and development in artificial intelligence. The government has also made specific policies to guide financial institutions to provide credit, promoting social capital to solve the common financing problems of enterprises. Third, the Chinese government developed the strategy of industry-university synergy. By encouraging universities, research institutes and enterprises to build innovation platforms, the transformation of technology research results would be accelerated. The strategy also involves building industrial parks, pooling resources and forming the advantages of industrial clusters. Last but not least, the government also supports the talent strategy. The relevant departments optimize the education system, from basic education to professional training in colleges and universities. The government introduced preferential policies to attract domestic and foreign talents to join the AI field as well.

2.2 Policies and regulations on data protection

Building artificial intelligence is a complex iterative process that includes data collection, model selection, parameter tuning, and validation. Among them, data processing is the core link that determines the success or failure of the model, but it is difficult to ensure data security by enterprise self-discipline alone. China has established a regulatory framework through legislation to balance technological innovation and privacy protection.

The 2017 Cybersecurity Law established basic rules for data security for the first time, requiring operators to adopt encryption measures in data collection, storage and transmission, laying the foundation for AI data security. Since AI was not yet widespread, the law did not provide enough protection for specific domains. In 2021, the Data Security Law established a hierarchical protection system, requiring enterprises to implement differentiated protection according to the impact of data on national security, economy and society, and data in key areas should be protected at the top level of the whole life cycle. The Personal Information Protection Law, which took effect in the same year, established the "notice-consent" principle, which strictly limited the processing of sensitive information and gave users control. In addition, China has also introduced special regulations such as the Regulations on the Security Protection of Critical Information Infrastructure to specify the requirements of AI data protection from the level of technical specifications, forming a legal system that covers the whole process. Table 1 contains various regulations.

2.3 Supervision and enforcement mechanisms

As Liengpunsakul (2021) stated, in the artificial intelligence data protection regulatory system, multiple departments cooperate to maintain data security. The Cyberspace Administration of China is responsible for the content review and data compliance supervision of artificial intelligence applications on Internet platforms, such as reviewing the AI recommendation content of short video platforms, supervising the data collection of social platform advertising systems, and ordering rectification when problems are found to protect users' right to know and privacy. The Ministry of Industry and Information Technology focuses on regulating the data security management of telecom operators and Internet enterprises' artificial intelligence services, checking the transmission encryption of telecom operators' network fault diagnosis systems, and guiding Internet enterprises to improve the storage security of smart home AI control systems. The State Administration for Market Regulation mainly maintains the order of market competition, regulates the data management behavior of enterprises, handles unfair competition cases such as e-commerce platforms using AI to grab competitors' data, corrects exaggerated propaganda of intelligent voice assistants, and protects consumers' rights and interests. The Ministry of Public Security has cracked down on artificial intelligence data crimes, taken quick action to arrest cases such as hackers stealing data from financial institutions, and carried out special operations to use AI to crack down on illegal collection and sale of data in the black and gray industry, and purify the network environment.

However, there are obstacles to multi-sector collaboration in the regulatory process as Oseni (2021) stated. First, there are drawbacks in the information sharing mechanism between departments. When investigating artificial intelligence data abuse cases, different departments have different access to data channels and standards. For example, in the data abuse case of an AI marketing company, the Cyberspace Administration of China, the State Administration for Market Regulation, and the Ministry of Public Security collect data in different ways, formats, and concerns, which leads to time-consuming and laborious data integration, delays in the investigation progress, and may miss key evidence, affecting the efficiency of law enforcement and regulation. Second, the division of responsibility is fuzzy. In emerging AI application scenarios such as intelligent driving data supervision, the responsibilities of the traffic management department and the network and information department have intersection, and when intelligent driving data leakage occurs, the two sides shirk each other, which expands the impact. The root cause is that the integration of emerging AI technologies breaks traditional industry boundaries, and existing regulations and departmental responsibilities have not kept up in time, hindering the development of the industry and damaging public trust in new technologies.

Given these challenges, improvements are urgent. On the one hand, a cross-departmental data sharing platform is established to learn from the experience of the European Union, and China takes the lead in organizing various regulatory authorities to formulate unified data standards for artificial intelligence data supervision and standardize data interfaces. When investigating cases, various departments can quickly obtain data through the platform and connect seamlessly. For example, when investigating AI financial fraud cases, financial supervision, public security, network and information technology departments can synchronically obtain data, shorten the investigation cycle, and improve the accuracy and coordination of law enforcement. On the other hand, the boundaries of regulatory responsibilities of various departments in the emerging AI field should be clarified through legislation or the introduction of departmental coordination documents. Taking intelligent driving data supervision as an example, it is clear that the traffic management department is responsible for data supervision directly related to vehicle driving and traffic as Ding (2018) stated.

3 The necessity to Regulate AI Data Protection in China

3.1 International Data Protection Trends

As Alzahrani (2024) revealed, artificial intelligence is becoming increasingly critical, and administrations around the world are scrambling to control its development. The EU has introduced a variety of regulatory instruments, the core of which is the General Data Protection Regulation (GDPR). The regulation regulates online data protection and artificial intelligence in many ways. By setting strict desensitization procedures, enterprises are restricted from obtaining sensitive information. Table 2 reveals the difference of regulations between EU and China.

Taking Volkswagen as an example, the famous German automobile manufacturers. In the data collection of intelligent assisted driving, only data related to driving safety and performance optimization are collected, and personal information acquisition is constrained due to the limitation of data collection in GDPR. Moreover, if Volkswagen provides user driving data to third-party insurance companies for personalized insurance, it needs to re-evaluate and obtain user authorization; At the same time, enterprises should clearly explain the data collection, use and storage information to users, and the public will protect users 'right to know through product manuals, official websites and dealer channels.

As Saveliev (2021) stated, China has a huge digital market and a large number of AI companies. The scale and processing difficulty of administrative data are similar to those of the EU, and many companies are expanding overseas, especially in European markets. With different regulatory regimes in different regions, it is wise for Chinese companies to follow local management when operating overseas. Adopting GDPR-like principles enables data compliance and reduces risk.

Global awareness of data protection is growing, and regulations are becoming stricter. China's AI market is highly competitive and needs to meet international high standards to enhance competitiveness. AI products and services that meet international privacy standards are easy to win the trust of global users and open up international markets. Although China has advantages in data resources, there are imperfect data management systems, cross-border flow problems and regulatory enforcement gaps with the European Union. Therefore, China needs to learn from the experience of the EU, improve regulations according to the actual situation, and balance data protection and industrial development as Wu Ke (2021) stated.

3.2 Domestic Data Security Needs

On October 26, 2016, Ctrip issued an announcement of "upgrade security payment", saying that the payment page may not open normally during the upgrade. What seemed like a routine maintenance notice turned out to be the trigger for a data breach. After the release of the news, senior industry insiders acutely aware of Ctrip payment system anomalies, access delays, error prompts are not like the system upgrade can explain, all kinds of signs pointing to malicious hacker attacks.

Over the next few days, the picture became clearer. Hackers carefully planned to use advanced technology to find Ctrip online payment system vulnerabilities, speculated that the use of SQL injection vulnerabilities, bypassing firewall, intrusion detection system and other security mechanisms, directly hit the core database. After entering, hackers skillfully screen and steal key sensitive information such as user names, ID numbers, credit card numbers, payment passwords, and mobile phone numbers, and a large amount of data is illegally obtained, which seriously threatens the information security of users.

Users of all ages, regions and consumption levels are affected. After the incident was exposed, complaints, supervision and media arrived, and Ctrip fell into a crisis of trust. Facing the loss of users and reputation damage, Ctrip needed to invest a huge amount of money in public relations, compensating users and upgrading security systems. Users' privacy is violated, facing risks such as identity theft, telecom fraud, and fund theft, and their lives are disturbed or they may have economic losses. This incident exposed the vulnerability of domestic data security protection. Even though artificial intelligence is widely used and data is a key asset, there is no perfect artificial intelligence data protection standard, and the rights and interests of enterprises and citizens are easily damaged. Therefore, it is urgent to formulate standards to prevent data risks and ensure the healthy and stable development of the digital economy as Stahl (2018) stated.

3.3 Technical Development and Social Stability Needs

Keeping the right development of artificial intelligence is complicated. Data misuse hinders the healthy development of AI technology initially just as the discuss in Li Xiuquan (2021). In the data collection stage, excessive collection of data unrelated to AI applications will increase the cost of data processing, reduce the efficiency of the algorithm, and may also introduce noise data to affect the accuracy of the model. In the model training stage, if unauthorized or poor-quality data are used, the trained AI model may be biased. For example, the face recognition system has low recognition accuracy for specific skin color people due to the deviation of training data, which affects the promotion and application of the system.

On the other hand, the excessive collection of data especially those are not open to public or contain privacy content may also bring the enterprise with a great number of litigations a social dispute, which hinder its further development and improvement to a large extent as Dhirani (2023) stated. Some enterprises share user data with third parties for targeted advertising without user consent, and users frequently receive harassing advertisements, which causes user disgust. The phenomenon is commonly seen in many small-scale online enterprises lasting for many years. Up to now, many internet users can still receive a few meaningless ads and harassing phone calls and text messages from unknown telephone numbers. In serious cases, data leakage may lead to public resistance to AI technology and even the entire digital industry, destroying the atmosphere of social trust in emerging technologies and affecting social stability finally.

By reasons of the foregoing, regulating data protection plays a key role in guiding the healthy development of AI technology and maintaining social stability. As Wang Chao (2020) stated, reasonable data protection standards can promote enterprises to legally collect and use data, improve data quality, and optimize AI model performance. At the same time, relevant departments should protect citizens' privacy, enhance public acceptance and trust in AI technology, create a good social environment, and promote the harmonious development of AI technology and society.

4 China's Artificial Intelligence Data Protection Reform Initiative

4.1 Reform of data classification and classification protection

At present, with the vigorous development of artificial intelligence, China is actively exploring the practice of data classification and classification protection as Nassar (2021) stated. The development of key data catalogue is a key measure. Through combing massive data, it is clear that the data related to national security, economic lifeline and social stability can be controlled. For example, sensitive customer information and transaction data in the financial sector, national energy reserves data and critical facility operating parameters in the energy industry are covered. This initiative lays the foundation for data security and makes protection efforts more targeted.

As Wu Fei (2020) revealed, for different levels of data, matching the corresponding strength of security protection measures can improve the overall protection efficiency and reduce the risk. The reform of data classification protection is very important to ensure data security. From the perspective of protection, data classification makes security protection targeted, and high-level important data can be encrypted, stored and transmitted by cutting-edge encryption algorithms (such as quantum encryption technology if conditions permit). Combined with fingerprint, facial recognition and dynamic password, multi-factor identity authentication is used to strictly control access rights. Low-level generic data is secured with conventional security measures, such as common encryption algorithms and simple authentication methods. From the perspective of resource allocation, after clarifying the data level, enterprises and relevant institutions can reasonably allocate human, material, financial and other security resources according to the importance of data, focus on protecting high-level important data, avoid waste of resources, significantly improve the efficiency and flexibility of the overall data security protection system, and effectively reduce security risks.

4.2 Reform of management of cross-border data flows

With the rapid development of economic globalization and digital economy, cross-border data flow has become more and more frequent, which has become an important force to promote international cooperation and industrial development. China has established a comprehensive and rigorous cross-border data management policy system, in which the security assessment mechanism and the standard contract mechanism are very important as Saveliev (2021) stated.

Before data is exported, data processors should make a comprehensive and in-depth assessment of the outbound risks, taking into account the soundness and level of data protection regulations in the receiving country or region, as well as the type (sensitive personal data, trade secret data or ordinary data), scale (data volume, number of users involved) and other factors as Wu Wenjun (2020) stated. The standard contract mechanism provides a standardized and standardized agreement template, specifies the rights and obligations of the data provider and the receiver in data transmission, storage, use and other links, and specifies the security technical measures of the data receiver, the data leakage notification obligation and other measures to protect the rights and interests of the data subject.

From the perspective of cross-border data security, these mechanisms constitute a multi-level security protection system. The security evaluation mechanism screened risks before data leaving the country and intercepted high-risk transmissions. When data is transmitted, the standard contract mechanism relies on legal constraints to make the receiver comply with the data processing, so as to prevent the data from theft, illegal tampering, improper use as Zhu Jun (2018) stated.

In the international cooperation of artificial intelligence industry, enterprises need to invest more time, manpower, capital and increase operating costs to meet compliance requirements in the short term. But in the long run, strict management policies enhance data security trust in international cooperation. For example, when Chinese AI enterprises and foreign enterprises jointly develop, the foreign enterprises have more confidence in data security due to China's perfect cross-border data management policy, and promote in-depth cooperation between the two sides in technical exchanges and data sharing. The government will also promote broader international exchanges and cooperation in the AI industry in a safe and orderly framework, and help China's AI industry expand steadily in the global market as Zhong Yi (2021) stated.

4.3 Reform to protect the rights and interests of data subjects

However, China has always placed the protection of the rights and interests of data subjects at the core of its data protection work, and actively introduced a series of targeted reform measures as Ye, hongbiao(2024). The optimization of privacy policy is one of the important measures. In the past, the long and obscure privacy policy clauses of enterprises make it difficult for data subjects to understand how their data is processed. Nowadays, enterprises are required to explain the whole process information of data collection, use, storage and sharing to data subjects in a more clear, intuitive and understandable way. Many enterprises began to use visual means, such as making data processing flow chart, with different colors and graphics clearly marked data flow and processing mode; Or use plain language, avoid excessive jargon, and present common data processing questions and answers in a question-and-answer format. The establishment of a complaint and reporting mechanism provides an effective relief way for data subjects when their rights and interests are infringed. When the data subject finds that his/her right to know is not protected, such as the enterprise does not clearly inform the purpose of data collection; Or the right to choose is taken away, such as forcing users to agree to unreasonable data collection terms; Or the right to delete is blocked. When the enterprise fails to respond to the request for deletion in time, the data subject can give feedback to the relevant regulatory authorities or the internal supervision agency of the enterprise through specially established complaint and reporting channels, such as online complaint platforms, special complaint emails or telephone hotlines as Tucker (2018) noticed.

Many Internet lead enterprise have made drastic changes to the presentation of privacy policies due to the current needs. Some companies, such as google, have developed interactive privacy policy pages. After logging in, users can click on different data processing sections to view detailed information, and can consult relevant questions in real time to get instant answers. These reform measures effectively safeguard the legitimate rights and interests of data subjects and enhance public trust in the use of data. When the public has more control and security over their data rights and interests, they will be more willing to participate in digital economic activities and provide rich data resources for the development of artificial intelligence industry. At the same time, a good trust environment is also conducive to attracting more investment and talents into the AI industry, promoting industrial innovation and development, creating a healthy and sustainable development environment for the AI industry, and promoting the whole industry to move towards a more standardized, orderly and prosperous direction as Anyanwu (2024) stated.

5 conclusion

The essay focuses on artificial intelligence data protection in China. Artificial intelligence has become the core driving force for the development of global science and technology and economy. As its key assets, data protection issues cannot be ignored.

The essay first combs the development strategy of artificial intelligence in China, and the government actively promotes its development by means of legislation, tax incentives, industry-study-research collaboration and talent strategy. At the same time, the essay expounds the current situation of data protection, the Network Security Law and other related laws have built a regulatory framework, and multi-department cooperation to maintain data security, but there are problems such as poor information sharing and fuzzy division of responsibilities between departments. Secondly, it analyzed the necessity of regulating data protection. With the increasingly strict global data protection in the world, China should learn from the experience of the European Union to improve the competitiveness of the AI market. On the domestic side, Ctrip data breaches and other incidents have exposed data security vulnerabilities, and data abuse hinders technological development and affects social stability. Finally, a number of reform measures were put forward, in the data classification and classification protection, the key data directory was formulated, and the security measures were matched according to the data level. Cross-border data flow management to build security assessment and standard contract mechanism; The reform of data subject rights protection optimizes the privacy policy and establishes a complaint and reporting mechanism.

There are shortcomings in current AI data protection work at the same time. When multi-department supervision, the information sharing mechanism is defective, different departments have different channels and standards for obtaining data, and the integration of data is time-consuming and laborious, which greatly affects the efficiency of law enforcement. In addition, the division of responsibility is vague, and in emerging AI application scenarios, departments are prone to shirk each other, hindering the development of the industry. In addition, although China is rich in data resources, its data management system is not perfect, and there is a regulatory and enforcement gap between China and the European Union in cross-border data flow, which makes Chinese AI companies face compliance challenges in the international market.

However, it’s necessary to improve the strategy. First of all, the government should do its best to formulate special AI data protection regulations, refine existing legal provisions, and clarify legal responsibilities to fill the legal gap. Secondly, relevant department can establish technology evaluation mechanisms, promote the application of legal science and technology, make the law adapt to the development of technology, and improve the efficiency of supervision. Thirdly, enterprise should be force to explore the establishment of data sharing norms, strengthen the application of privacy enhancement technology, and realize the balance between data value and privacy protection. Fourth, both government and organizations should promote international cooperation, formulate unified standards, and establish regulatory platforms to address regulatory challenges for cross-border data flows, and ensure cross-border data security. Last but not least, relative department can carry out publicity and education, promote the development of industry self-discipline and norms, raise public awareness of data protection, and strengthen corporate social responsibility.

Looking forward to future research directions, the government should formulate special AI data protection regulations, refine existing provisions, clarify legal responsibilities, and fill legal gaps. We will establish a technology evaluation mechanism, promote the integration of law and science and technology, and improve regulatory efficiency. Enterprises need to explore data sharing norms and apply privacy enhancement technologies to balance data value and privacy protection. The government and relevant organizations should strengthen international cooperation, formulate unified standards and build supervision platforms to ensure cross-border data security. Publicity and education activities should also be carried out to promote industry self-discipline and enhance public awareness of data protection and corporate social responsibility.

reference

Allen, Gregory C. "Understanding China's AI strategy: Clues to Chinese strategic thinking on artificial intelligence and national security." (2019).

Alzahrani, Reema Bakheet. "An overview of AI data protection in the context of Saudi Arabia." International Journal for Scientific Research 3.3 (2024): 199-218.

Anyanwu, Evangel Chinyere, et al. "Artificial intelligence in healthcare: a review of ethical dilemmas and practical applications." International Medical Science Research Journal 4.2 (2024): 126-140.

Dhirani, Lubna Luxmi, et al. "Ethical dilemmas and privacy issues in emerging technologies: A review." Sensors 23.3 (2023): 1151.

Ding, Jeffrey. "Deciphering China's AI dream." Future of Humanity Institute Technical Report (2018).

Gao, zhao-shan, et al. "AI in China: a survey." IEEE Intelligent Systems23.6 (2008): 26-32.

Li Xiuquan. "Security, privacy and ethical challenges in the application of artificial intelligence and coping thinking." Science and Technology Guide 35.15 (2017):2.

Li, xiuquan, and Tao Zhang. "An exploration on artificial intelligence application: From security, privacy and ethic perspective." 2017 IEEE 2nd international conference on cloud computing and big data analysis (ICCCBDA). IEEE, 2017.

Liengpunsakul, Subin. "Artificial intelligence and sustainable development in China." The Chinese Economy 54.4 (2021): 235-248.

Nassar, Ahmed, and Mostafa Kamal. "Ethical dilemmas in AI-powered decision-making: a deep dive into big data-driven ethical considerations." International Journal of Responsible Artificial Intelligence 11.8 (2021): 1-11.

Oseni, Ayodeji, et al. "Security and privacy for artificial intelligence: Opportunities and challenges." arxiv preprint arxiv:2102.04661 (2021).

Saveliev, Anton, and Denis Zhurenkov. "Artificial intelligence and social responsibility: the case of the artificial intelligence strategies in the United States, Russia, and China." Kybernetes 50.3 (2021): 656-675.

Stahl, Bernd Carsten, and David Wright. "Ethics and privacy in AI and big data: Implementing responsible research and innovation." IEEE Security & Privacy 16.3 (2018): 26-33.

Tucker, Catherine, et al. "Privacy, algorithms, and artificial intelligence." The economics of artificial intelligence: An agenda (2018): 423-437.

Wang, Chao, et al. "Privacy protection in using artificial intelligence for healthcare: Chinese regulation in comparative perspective." Healthcare. Vol. 10. No. 10. MDPI, 2022.

Wang, Ke, et al. "Analyzing the adoption challenges of the Internet of things (Iot) and artificial intelligence (ai) for smart cities in china." Sustainability 13.19 (2021): 10983.

Wu, Fei, et al. "Towards a new generation of artificial intelligence in China." Nature Machine Intelligence 2.6 (2020): 312-316.

Wu, Wenjun, Tiejun Huang, and Ke Gong. "Ethical principles and governance technology development of AI in China." Engineering 6.3 (2020): 302-309.

Ye, hongbiao, et al. "Privacy and personal data risk governance for generative artificial intelligence: A Chinese perspective." Telecommunications Policy 48.10 (2024): 102851.

Zhang Huilian,and Wang Cheng." The Ethical Dilemma of Generative AI Technology Transition and Its Resolution." Journal of Hefei University of Technology (Social Science Edition) 38.05(2024):51-58.

Zhang, Yi, et al. "Ethics and privacy of artificial intelligence: Understandings from bibliometrics." Knowledge-Based Systems 222 (2021): 106994.

Zhu, Jun, et al. "The future of artificial intelligence in China." Communications of the ACM 61.11 (2018): 44-45.