In late September 2024, the State Counsel published the Administrative Regulation on Network Data Security (“Regulation”), which came into effect on 1 January 2025. The Regulation establishes a comprehensive framework governing the security of a wide range of data, including personal information (“PI”) and important data. It also introduces a series of obligations for network data processors and strengthens regulatory oversight.
This is the second article of our series. In this article, we outline the key provisions of the Regulation, with a particular focus on the differentiated requirements for various categories of data, and share our observations on its potential implications.
If you would like to subscribe for our newsletters and be notified of our events on China cybersecurity and data protection, please contact James Gong at james.gong@twobirds.com.
China’s data protection framework is structured around three fundamental pillars: the Cybersecurity Law (“CSL”, 2017), the Data Security Law (“DSL”, 2021), and the Personal Information Protection Law (“PIPL”, 2021). While these laws overlap in certain areas, each focuses on a distinct dimension of data governance.
The CSL primarily regulates the security of networks and information systems. At the time of its enactment, in the absence of a dedicated important data and PI protection law, it also contained some initial provisions on PI protection and mentioned important data. The DSL provides a general regulatory framework for all types of data but places particular emphasis on safeguarding national security and public interests. The PIPL, which came into force most recently, establishes a comprehensive regime for the protection of PI and serves as China’s equivalent to the GDPR.
Although these three laws are interconnected, their respective areas of focus are distinct. The Regulation aims to consolidate and refine the rules set out in these foundational laws in the specific context of network data. It provides greater clarity by addressing the differentiated requirements for general network data, PI, and important data, thereby further strengthening China’s data governance regime.
The Regulation sets out comprehensive obligations for network data processors across three key areas - general network data, PI, and important data. It outlines baseline security and compliance requirements applicable to all data, establishes dedicated rules for PI processing, and introduces enhanced protections for important data with national security implications. In the following sections, we examine these requirements in greater detail.
Chapter II of the Regulation outlines the general provisions applicable to all network data processing activities, regardless of the type of data involved. In contrast to the more specific requirements for PI and important data under Chapters III and IV, the provisions under Chapter II establish baseline security and compliance standards for all network data processors. As mentioned in the first article of this series, the definitions of network data and network data processing activities are very broad, so the scope of application of Chapter II is also extensive. The general obligations can be summarized as follows:
1. Prohibition of Using Network Data for Illegal Activities
Network data processors must not use network data to engage in illegal activities. They should not illegally steal or obtain network data, nor illegally sell or provide network data to others. Additionally, no individual or organization may provide programs, tools, technical support, advertising, payment settlement, or other assistance for such illegal activities.
The draft version of Regulation provided a detailed enumeration of specific types of illegal activities. However, in the enacted version, the exhaustive list of illegal activity types was not retained. Instead, the prohibition of illegal activities is articulated in more general terms. This change may have been made in consideration of the fact that, as technology evolves, the forms of illegal activities are likely to become more diverse.
Furthermore, the Regulation clarifies that prohibited technical support for illegal activities includes internet access, server hosting, network storage, and communication transmission, extending the obligations to certain technical service providers.
2. Network Security Responsibility
The Regulation requires network data processors to bear the primary responsibility for network data security, strengthening the security protection of network data through institutional and technical measures on the foundation of multi-level network security protection. If network data needs to be transferred due to mergers, divisions, dissolution, or bankruptcy, the receiving party must continue to fulfil the obligations of network data protection and bear the primary security responsibility.
In addition, network data processors must also pay attention to the following obligations related to network data security:
Notably, compared to the DSL and the PIPL, the Regulation introduces several new requirements: (i) it explicitly mandates the execution of a DPA for the provision, entrusted processing, and joint processing of important data, which is not required under the DSL; (ii) it requires a DPA to be signed when providing PI to another processor, a requirement not clearly stipulated under the PIPL; and (iii) it emphasizes the obligation to retain records of the processing of PI and important data for at least 3 years. While this requirement exists under Article 55 of the PIPL, it has not been effectively implemented in practice, as Article 55 primarily focuses on conducting PI protection impact assessments rather than on record-keeping.
The Regulation provides more detailed guidance on PI protection based on the requirements set forth by the PIPL, mainly focusing on notification, consent, data subject rights, and data retention.
1. Notification
With respect to the notification requirements, before processing PI, network data processors must develop and publicly display PI processing rules that are easily accessible, prominently positioned, and clearly written. These rules must include at least the following:
Compared with the PIPL’s requirements, the Regulation further emphasizes that if the retention period is difficult to determine, the method for determining the retention period should be clearly indicated.
To ensure individuals can clearly understand the details of processing activities, the Regulation also requires network data processors to list the details of the processing and collection in the form of a checklist or similar format.
Note: Network data processors are advised to provide clear lists of the types of PI collected and the lists of third parties, including Software Development Kits (SDK) providers, with whom such information is shared.
2. Consent
Consent is the most important and commonly used legal basis in China. The Regulation, addressing this important legal basis for data processing under the PIPL, further clarifies the following:
3. Data Subject Rights
To ensure users can conveniently exercise their rights related to PI as stipulated by the PIPL, the Regulation requires network data processors to promptly handle requests and provide convenient methods and channels to support individuals in exercising their rights. Unreasonable conditions that restrict legitimate requests from individuals should not be imposed.
Specifically, for right to data portability, network data processors should provide access and retrieval methods for the PI to the other network data processors designated by the individual, if the following conditions are met: (1) The requester's true identity can be verified; (2) The information to be transferred is either provided with the individual's consent or collected based on a contract; (3) The transfer of PI is technically feasible; (4) The transfer of PI does not harm the legitimate rights and interests of others.
4. Other important issues
The PIPL remains silent on the methods and channels for reporting the contact information of local representatives designated by foreign PI processors under its extraterritorial effect. While the Regulation also do not establish such a channel, it indicates that the contact details of the local representatives should be provided to the municipal-level CAC.
Based on a risk-based approach, network data processors handling PI of over 10 million individuals must also comply with certain provisions of the Regulation applicable to processors of important data. This includes designating a person responsible for network data security and establishing a dedicated network data security management agency. They must fulfil relevant network data security protection responsibilities and report important data handling plans to provincial-level or higher authorities, especially in cases of mergers, splits, dissolution, or bankruptcy that may affect data security.
Chapter IV of the Regulation outlines the provisions that network data processors processing important data, or important data processors, must comply with. As mentioned in the first article of this series, the important data is defined under the Regulation as data that (i) if modified without permission, destroyed, leaked, or illegally acquired or used, may directly harm national security, economic operation, social stability, and public health and safety and (ii) within specific sectors, groups or regions or reaching a certain level of precision or scale.
The obligations that important data processors must comply with can be summarized as follows:
1. Identification and declaration of important data
For a long time, identifying important data has been a significant challenge for enterprises. To tackle this issue, Article 29 of the Regulation introduces a three-step mechanism:
However, to date, most sectoral regulators and regional authorities have not released catalogues or identification rules for important data[1], nor have they provided detailed procedures for the declaration of important data. (Please see the first article of this series for more information about the definition of important data) Therefore, although the Regulation clearly mandates that network data processors must identify and declare important data, the specific implementation details still await further clarification from competent authorities.
2. Implementing important data security measures
The Regulation requires that important data processors must designate a network data security officer and establish a network data security management institution. Additionally, in cases of mergers, splits, dissolutions, or bankruptcies, they must report the important data disposal plan to the competent authorities.
3. Conducting important data risk assessments
The Regulation requires that important data processors must conduct (1) annual risk assessments of their network data processing activities and (2) risk assessments before providing, entrusting, or jointly processing important data. The focus of these two types of assessments differs.
Unlike the annual risk assessment, the Regulation does not require this type of assessment to be submitted to competent authorities. Additionally, this assessment focuses more on evaluating the legality, legitimacy, necessity, associated risks, and security measures of the provision, entrustment, or joint processing of important data, as well as assessing the integrity and lawfulness of the data recipient.
The Regulation marks a significant step forward in China’s data governance landscape by refining the protection of PI and important data. It strengthens the enforcement of existing laws by clarifying network data processors’ obligations in areas such as consent management, individual rights protection, and security requirements. Notably, for important data, the Regulation establishes a basic regulatory framework.
From a compliance perspective, enterprises should revisit their existing PI protection programs, even if they have already implemented measures under the PIPL. The Regulation imposes additional requirements in areas such as obtaining valid consent, responding to data subject rights requests, executing DPAs for PI provision, and retaining records of PI transfers. Companies should carefully review these new obligations to identify any compliance gaps and make targeted adjustments.
With respect to important data, although the Regulation outlines a preliminary governance structure, many industries have yet to publish detailed important data catalogues or sector-specific regulatory rules. In light of this, enterprises are advised to closely monitor future developments within their respective industries and regulatory sectors and to prepare for compliance efforts once further guidance is issued.
[1] Currently, only a few sectoral regulators and regional authorities have issued catalogues or identification rules for important data. For example, the free trade zones in Beijing, Shanghai, Tianjin, and Fujian have released some negative lists or general data lists for cross-border data transfer, which include catalogues for identifying important data. Additionally, the automotive sector regulators have issued the rules for identifying important data in the automotive sector.