Suitable for all audiences? [Part I]

Four years ago, China revised its Law on the Protection of Minors, limiting the visibility of online content deemed potentially harmful to children, and directing regulators to clarify the scope of that broad category at a later stage. These rules are important beyond the stated goal of child-protection, because barring such content from prominent spaces, such as website homepages and lists of trending topics, inevitably reduces adults’ exposure to it as well. What’s worse, the threat of steep penalties for violations gives platforms strong incentives to err on the side of caution by pushing ever more information to the sidelines.

Earlier this year, long-anticipated clarification finally came into sight with the release of draft guidance for identifying this potentially harmful content. Unfortunately, the new authority does little to resolve the uncertainty. The draft remains vague, internally inconsistent, and misaligned with the existing law; creating as many problems as it solves.

Background:

The new draft Measures for Categorizing Online Information that Might Impact Minors’ Physical and Psychological Health  can only be understood in the context of China’s broader information controls.

China’s general online content control regime focuses on three broad groups of content: encouraged ‘positive content’, prohibited ‘illegal content’, and discouraged ‘negative content’. [see table 1 below]

It is a punishable offense to create, reproduce, or transmit “illegal content”, such as information promoting terrorism, attacking the national leadership, or leaking state secrets. The broader category of “negative content” is generally permissible, but platforms can still be punished for actively promoting it by placing it in certain key locations. Those high-visibility areas should instead be filled with neutral, or ideally, “encouraged” positive-energy content.

Table 1: Summarizing the main categories of regulated content.

Content for Minors

The revised Law on the Protection of Minors (LPM) and subsequent Regulations on the Protection of Minors Online (RPMO) addressed two additional categories of content: (1) Content harmful to minors, and (2) content that “might impact minors’ physical or psychological health.”

The first category, content harmful to minors, includes prohibited “illegal content” from the broader regulatory scheme, as well as content promoting self-harm, and is thus largely prohibited for minors and adults alike.

Online platforms, game designers, users, and parents have special obligations to help minors avoid the second category of information, which is referred to here as “potentially harmful to minors” (PHM) content. The obligations primarily include:

• [No Targeting Minors] This content must never appear in online products and services that are specifically for minors, and must not be sent directly to minors [LPM 80, RPMO 24, 25]
• [Labeling Requirement] Where such content does appear in services or products, it must be clearly labeled as potentially harmful to minors, and platforms are responsible for ensuring that users also appropriately label their content. [LPM 51, RPMP 23, Draft article 9]
• [No Promotion] Like “negative content” in the general content regulation scheme, PHM content must never appear in eye-catching positions, such as home pages, pop-up windows, or trending topic lists. [RPMO 24]

As mentioned above, the LPM left defining the specific scope of PHM content to future regulations. The Regulations on the Protection of Minors Online only added that PHM information includes content that might lead minors to imitate unsafe behaviors or violate social mores, or that could cause extreme feelings, or lead to bad habits. This description was also included in recent guidelines on “minors’ modes for mobile internet,” and with the exception of ‘extreme feelings’ reflects “negative content” items under the broader content regulations.

Terms like “unsafe behaviors”, “violating social mores”, “extreme feelings”, and “bad habits” read like a wish list of stereotypical adolescent traits the regulators might hope to curb, but are too vague to serve as workable standards for content regulation.

The New Draft

The Draft Measures are the long-overdue response to the Law on the Protection of Minors’ call, made four years earlier, for the cybersecurity authorities to work with other departments to determine the types and scope of PHM content. Unfortunately, they fail to deliver the promised clarity.

The Measures begin with a definition of PHM content in Article 2 that essentially recites the categories of PHM previously raised by the Regulations on the Protection of Minors Online. When the Measures continue by providing detailed illustrative lists of individual types of PHM content in articles 3-7, however, they don’t closely follow that definition.

The first two categories, information causing minors to imitate negative conduct or impacting minors’ values, loosely correspond to the Measure’s definition of PHM. Content encouraging imitation of unsafe conduct or causing extreme feelings is folded into the first category as examples of ‘negative conduct’, as are more specific bad habits, such as smoking, drinking, and celebrity worship. The following three categories, however, go in a very different direction, discussing information using minors’ images, disclosing minors’ personal information, or leading to internet addiction, and this shift creates a few problems.

The first issue is likely just the result of sloppy drafting. Not including the three new categories in the formal definition of PHM content creates a basic disconnect: the subsequent lists elaborating on those categories can’t help clarify the definition if they illustrate terms that aren’t even a part of that definition at all.

The second issue is a more fundamental conceptual disconnect. The potential harm in PHM content comes from minors’ exposure to certain information. Accordingly, the rules for handling PHM content are all aimed at helping minors avoid that harmful content, such as by clearly labeling it and removing it from prominent positions and recommendations. The definitions’ categories of PHM are identified solely by the negative effects they may have on minors, such as leading them to imitate bad conduct or triggering extreme emotional responses.

The new categories, however, don’t follow this logic. Instead, they discuss information which may be harmful to minors in other ways, such as in its production (as in short videos pranking or mocking minors) or from other people’s access to and abuse of the information (as in leaks of personal information). Labeling such information as harmful to minors and limiting minors’ exposure to this information won’t prevent the harm involved.

China has separate legal authority aimed expressly at protecting minors’ personal information, fighting internet addiction, and stopping the online financial and sexual exploitation of minors. These issues are further addressed in the comprehensive Regulations on the Protection of Minors Online.  Including them as PHM isn’t helpful.

In addition to these new problems, the original problem of the rules being unworkably vague remains. How should a platform or user identify, for example, content that might “entice minors to make bad/undesirable friends” [不良交友的].  Consider Table 3’s summary of each category in the draft.

Table 3: Overview of Information that Might Impact Minor’s Health in New Draft

Protection of minors online is an area that China has been taking very seriously for some time, and the draft is hopefully a misstep that will be corrected before being finalized.

In determining the appropriate scope of PHM content, regulators need to keep in mind the incidental impact on adults from sidelining this content and the burden on users and platforms in labeling all such content. At the same time, the definition needs to recognize and align with existing legal authority, including that already requiring that “negative content” be kept out of highly-visible areas.

The current framework seems to connect”content harmful to minors” with “illegal content” under the broader content regime, and PHM content with “negative content”, but stops short of fully equating them. If the distinction is to continue with regard to online content, regulators will have to explain what makes PHM more hazardous to minors than other negative content.

Jeremy Daum

Jeremy Daum is a Senior Fellow of the Yale Law School Paul Tsai China Center, based in Beijing, with over a decade of experience working in China on collaborative legal reform projects. His principal research focus is criminal procedure law, with a particular emphasis on protections of vulnerable populations such as juveniles and the mentally ill in the criminal justice system, and is also an authority on China’s ‘Social Credit System’. Jeremy has spoken about these issues at universities throughout China and in the U.S.; and has co-authored a book on U.S. Capital Punishment Jurisprudence for Chinese readers. He is also the founder and contributing editor of the collaborative translation and commentary site Chinalawtranslate.com, dedicated to improving mutual understanding between legal professionals in China and abroad.
He translates, writes, edits, does web-design, graphic design, billing, tech support, and social media outreach for China Law Translate.

• Jeremy Daum

  

  2026/05/14

  Tong Lihua and the “Law of the User”: A User Rights’ Based Approach to Platform Regulation
• Jeremy Daum

  

  2026/04/13

  The Friends We’ve Made 2: Changes from the Draft
• Jeremy Daum

  

  2026/04/09

  Digital Humans vs. Human-like AI
• Jeremy Daum

  

  2025/12/28

  The Friends We’ve Made: An Initial Breakdown of Draft Rules for Human-like AI