AI备案

… The Chinese regulators adopt a holistic … of algorithm with self-assessment reports through a state filing system. Critics might point out the risk of code censorship via this algorithm filing …

ABSTRACT This article maps the trajectory of China’s regulation of algorithms via policy review. It divides China’s governing progress into three phases: the ‘post-event policy response and penalty’ phase, the ‘ethics guidelines, guiding opinions and self-discipline pacts’ phase and the ‘legislation and implementation’ phase. The paper argues that the ideological and political implications of algorithmic applications are the highest concern for Chinese regulators. China’s regulation of algorithms follows a ‘state-centric multilateral model’ – the same model used for its internet governance. The ‘algorithmic transparency’ advocated by regulators is currently only limited to algorithms in the platform economy and industries rather than those used for government decision-making and public administration. As the first nation to issue laws regulating algorithms and generative AI, China faces problems and challenges emerging from further implementing the laws. China’s experience will provide valuable first-hand understandings for countries currently creating legal frameworks to regulate algorithms and AI.

The People’s Republic of China (PRC) has moved ahead of other jurisdictions in regulating Artificial Intelligence (AI) by enacting a set of regulations aimed at addressing risks associated with AI. This article reviews three key AI specific regulations enacted in the PRC in 2022 and 2023. Together these regulations set out comprehensive obligations on the use of generative AI and algorithmic recommendation services in the PRC. The obligations include general requirements such as to comply with the law and principles of fairness as well as to protect against incorrect political direction and violation of social morality. There are also more specific obligations to achieve those objectives. These include, ensuring the regulator has visibility of certain algorithms by requiring the filing of information about the relevant algorithms. Further there are specific obligations on service providers such as to label training data appropriately, take measures to screen illegal or harmful content, verify the identities of users for some services, protect against fake news and label artificial intelligence generated content as synthetic. The regulations also provide a suite of rights for users, including the right not to be targeted on the basis of their personal characteristics by services using AI algorithms. The obligations under these regulations are significant and it will be interesting to see how the PRC’s regulatory approach impacts the AI industry in the PRC in contrast to other jurisdictions where regulators are still looking at how to regulate AI or taking a different approach. PRC, China, Artificial Intelligence, AI Regulations

In China, algorithms have been increasingly used in many different sectors to facilitate analysis of massive data and optimise the decision-making process. While this approach brings significant benefits, the complex design of algorithmic models and the large scale of data involved pose serious challenges to the existing regulations. In response, China has gradually established a regulatory regime covering many areas of law, such as rules governing personal information protection and algorithm recommendation services, to oversee the development and use of algorithmic decision-making. However, China’s regulatory regime is not without its limitations. Drawing on the regulatory experience of overseas jurisdictions, including the EU, the US, the UK and Singapore, this paper makes some suggestions for improving Chinese regulations. It is worth considering formulating specific requirements for data used in algorithmic decision-making and complementing the existing regulatory regime with algorithmic audit mechanisms. China is also advised to strengthen private and public enforcement and incorporate the code of ethics into the governance structure of algorithmic decision-making.

… China has recently finalized its first algorithmic regulations—… or fuelling fake news via algorithms. Under the guidance of the … China's next regulatory target — Algorithms, the secret of …

… 74It should also be noted that in special individual-enterprise areas such as finance, social media and mega-platform management, a special algorithm filing system was created based …

… Internet information service algorithm filing system, and go through the filing procedures. … algorithm filing system, information filled in by enterprises mainly includes three parts: algorithm …

… Union, and China choose to regulate recommendation algorithms will undoubtedly have a … Compliance with the regulation also requires the filing of an AI self-assessment report (…

… , China's new rules on generative artificial intelligence (AI) … context for this rapid regulatory development. It argues that … of China's attempt to adopt tailored rules for generative AI, and the …

… The registry includes requirements to disclose the sources of training data, and the draft generative AI regulation has specific requirements on the data’s diversity and “objectivity.” Many …

… -Regulation model could be an ideal alternative for China. … ‘Meta-Regulation’ aligns with the generative AI governance … -Regulation theory, the governance of generative AI in China …

… The way the great powers of the internet — the United States, the European Union, and China — choose to regulate generative AI will undoubtedly seriously impact our lives and politics…

The rapid development of generative artificial intelligence (AI) systems, particularly those fuelled by increasingly advanced large language models, has raised concerns of their potential risks among policymakers globally. In July 2023, Chinese regulators enacted the Interim Measures for the Management of Generative AI Services (“the Measures”). The Measures aim to mitigate various risks associated with public-facing generative AI services, particularly those concerning information content safety and security. China’s approach to regulating AI to date has sought to address the risks associated with rapidly advancing AI technologies while fostering innovation and development. Tensions between these policy objectives are reflected in the provisions of the Measures. As Beijing moves towards establishing a comprehensive legal framework for AI governance, there will be growing interest in how China’s approach may influence AI governance and regulation at a global level.

This study examines policy interventions and regulation of generative artificial intelligence (AI), focusing on key differences in generative AI policy in the EU, the US, and China. Using a comparative research methodology, the study analyzed the most recent policy documents from these regions through text-mining techniques to assess their key differences in terms of word frequency and specific content. This work highlights the different strategies, goals, and approaches to regulating generative AI across the regions. It found that the EU adopts a more comprehensive and stringent regulation of generative AI, emphasizing regional harmonization and foresight; the U.S. regulation is characterized by pragmatism, closely aligned with industry innovation, and a focus on risk avoidance; while China focuses more on macro-level regulation aimed at promoting innovation and encouraging ecological construction. Participants may be interested in this study because it not only uses up-to-date materials but also employs text-mining methods to present the findings in a clearer way than previous studies. It provides insights into how regulatory policies for generative AI affect the development and practice of digital government. The study sheds light on different countries’ strategies for technology governance, which is crucial for understanding and designing effective digital government policies. In addition, due to the potential of generative AI technologies to deliver public services and drive government transparency, these insights help participants better assess the opportunities and challenges of technological innovation in the digital government space.

Conroy, G. 2023. Scientists used ChatGPT to generate an entire paper from scratch—but is it any good? Nature 619 (7970):443–4. doi: 10.1038/d41586-023-02218-z. Dastin, J. 2018. Amazon scraps secret AI recruiting tool that showed bias against women. Reuters, October 10. https://www.reuters.com/article/us-amazon-com-jobsautomation-insight-idUSKCN1MK08G Gold, E. R. 2021. The fall of the innovation empire and its possible rise through open science. Research Policy 50 (5): 104226. doi: 10.1016/j.respol.2021.104226. Hosseini, M., C. A. Gao, D. M. Liebovitz, A. M. Carvalho, F. S. Ahmad, Y. Luo, N. MacDonald, K. L. Holmes, 2023. An exploratory survey about using ChatGPT in education, healthcare, and research. MedRxiv: The Preprint Server for Health Sciences doi: 10.1101/2023.03.31. 23287979. Hosseini, M., D. B. Resnik, and K. Holmes. 2023. The ethics of disclosing the use of artificial intelligence in tools writing scholarly manuscripts. Research Ethics doi: 10.1177/ 17470161231180449. Jones, D. 2021. Facebook apologizes after its AI labels black men as ’primates’. NPR, September 4. https://www.npr. org/2021/09/04/1034368231/facebook-apologizes-ai-labelsblack-men-primates-racial-bias Lipenkova, J. 2023. Overcoming the limitations of large language models: How to enhance LLMs with human-like cognitive skills. Towards Data Science, January 23, https://towardsdatascience.com/overcoming-the-limitations-of-large-language-models-9d4e92ad9823 Lubell, J. 2023. ChatGPT passed the USMLE. What does it mean for med ed? AMA, March 3. https://www.ama-assn. org/practice-management/digital/chatgpt-passed-usmlewhat-does-it-mean-med-ed National Academies of Science, Engineering, and Medicine. 2017. Fostering Research Integrity. National Academies Press, Washington, DC. Park, M., E. Leahey, and R. J. Funk. 2023. Papers and patents are becoming less disruptive over time. Nature 613 (7942):138–44. doi: 10.1038/s41586-022-05543-x. Porsdam Mann, S., B. D. Earp, N. Møller, S. Vynn, and J. Savulescu. 2023. AUTOGEN: A personalized large language model for academic enhancement—ethics and proof of principle. The American Journal of Bioethics 23 (10):28–41. doi:10.1080/15265161.2023.2233356. Shamoo, A. E., and D. B. Resnik. 2022. Responsible conduct of research. 4th ed. New York: Oxford University Press. Weiss, D. C. 2023. Latest version of ChatGPT aces bar exam with score nearing 90th percentile. ABA Journal, March 16. https://www.abajournal.com/web/article/latestversion-of-chatgpt-aces-the-bar-exam-with-score-in-90thpercentile

: Generative artificial intelligence technologies, represented by ChatGPT, have achieved widespread applications that have transformed traditional interaction patterns between humans and artificial intelligence and have had profound impacts on social production methods. Both domestically and internationally, academic discussions have emerged on the topic of artificial intelligence agency and legal regulations to address a range of challenges posed by generative artificial intelligence, including ethical and moral issues, intellectual property protection, privacy and data protection, market monopolies, cybercrime, and data security. At the current stage, China should adhere to a balanced approach that emphasizes both security and development in the governance of generative artificial intelligence. Based on the principle of placing people at the center, we should promote the establishment of an artificial intelligence ethics code and promote the development of a systematic legal regulatory system that is founded on general generative artificial intelligence legislation and supplemented by specific management measures.

… This article provides a detailed description of China’s regulatory ideas and specific methods… generative AI in China. This article analyzes the governance path of generative AI in China …

… not only make it more difficult to regulate AI but also present challenges for the current legal … personal data protection in China. This article finds that while China's governance on privacy …

Objective: to trace the development trajectory of legal regulation in the field of artificial intelligence in the People’s Republic of China by revealing the advantages and disadvantages of China’s approach to artificial intelligence regulation and to outline the prospects of national regulation for the nearest future, taking into account the world experience.Methods: general scientific methods of analysis and synthesis, classification, systemic and functional approaches. Also, the formal-legal, comparativelegal, and historical-legal methods were used.Results: the research demonstrates the validity of Chinese claims for world leadership in the creation of legal regulation of artificial intelligence, as it is in China that the first normative legal acts were adopted. These acts have already entered into force; however, each of them deals with a narrow range of issues, while there is no law to establish general rules for the artificial intelligence industry. Among the characteristic features of the Chinese approach we can name, first of all, its iterative nature, which allows adjusting the regulation with each new step. Another feature is the sectoral nature of the regulation.Scientific novelty: in the course of the research, the development stages of artificial intelligence legal regulation in China were identified and described; the advantages and disadvantages of the Chinese approach to regulation were identified and argued; this approach was compared with the approaches of China’s main rivals competing with it in terms of the technology development and its legal regulation. All of the above allowed making conclusions about the subsequent development of legal regulation in China and in the whole world.Practical significance: familiarization with the research materials enables interested legal scholars, and not only them, to get a clear idea of the level of artificial intelligence regulation, achieved by China. China’s experience is of significant interest to the rest of the world, showing the correctness or faults of possible regulatory options in the new and complex field. The study results can be used in the practice of legal regulation in the sphere of artificial intelligence, as well as in preparing lectures in the relevant courses and writing tutorials for law students.

The subject of the research is modern technologies of generative artificial intelligence (GII), their impact on society and law (using the example of China). The rapid development of GII is associated with the growth of venture capital investments and active support from large technology companies and states. Since 2022, China has adopted a number of laws on the regulation of artificial intelligence. At the same time, the PRC focuses on the unconditional protection of state security and national interests. An important aspect of AI regulation in China is the desire to form an AI bill that significantly expands the regulatory architecture. It is expected that the bill will be adopted during 2025, which will contribute to a more complete and detailed regulation of artificial intelligence. In the course of the research, the author used the following methods of cognition (research methodology): dialectical method of cognition, general scientific empirical methods of cognition (comparison and description), general scientific theoretical methods of cognition (generalization and abstraction, induction and deduction, analogy), as well as private scientific empirical methods of cognition (method of interpretation of legal norms) and private scientific theoretical methods cognition (legal and dogmatic). The main conclusions of the study are as follows. To date, the draft law on AI proposed by Chinese legal scholars is still under discussion, but it is already clear that it significantly complements and expands the already established architecture of legal regulation of artificial intelligence in the People's Republic of China. It contains a lot of bold ideas (for example, about the legal protection of data obtained as a result of the work of the GII). It seems that during 2025, the specified draft law (apparently with improvements) will be adopted. Based on the existence of regulatory legal acts that have already entered into force and are currently in force regarding artificial intelligence (including generative), as well as trends towards the rapid formation of the basic law on AI, it clearly follows that China is following the path of legal regulation of this area for general use within the PRC, while giving freedom of use and study AI for government purposes, in order to protect national interests.

Abstract The technological adoption and widespread use of generative AI, as represented by ChatGPT, have altered the conventional mode of interaction between humans and AI and profoundly changed the way our society operates. In response to the multifarious risks associated with generative AI, spanning ethical considerations, intellectual property protection, privacy and data protection, market monopoly, cybercrime, and data security concerns, discussions on the status of AI subjects and legal regulation have surfaced both on a global and domestic scale. It is argued in the present study that generative AI governance should uphold the principle of reconciling safety and development, craft an AI code of ethics in line with the umbrella of the human-centered principle, and promote a comprehensive legal framework of AI based on general AI legislation, supplemented by specialized management approaches for generative AI, and underpinned by established legal norms.