Online Reputation Governance Monthly Report (July 2026)

Theme of the month: the July storm over multinational brands' reputation in China — from the Driscoll's PFAS class action and the ZARA wide-leg trousers safety controversy to Louis Vuitton's RMB 10.3 million award against Molly Tea, a multi-dimensional legal response to product safety crises and trademark enforcement backlash. Includes analysis of six regulatory signals and counsel's prescriptions for two categories of scenarios.

Theme of the Month: The July Storm over Multinational Brands’ Reputation in China — A Multi-dimensional Legal Response, from Product Safety Crises to Trademark Enforcement Backlash

Three Key Takeaways This Month

Regulatory Signals This Month: Cyberspace regulators advanced simultaneously along three tracks — “legislation, special campaigns, and detailed rules.” The Anti-Cyberviolence Law of the People’s Republic of China (Draft for Comments) consolidates cyber-violence into dedicated legislation for the first time; the Measures for the Administration of Internet Information Services (Revised Draft) has been released for a second round of public comment to align with the Cybersecurity Law; the first phase of the “Qinglang · Rectifying AI Application Irregularities” campaign dealt with 14,000 AI products; the “short-video content labelling” drive removed 241,000 items of information; the “self-media information source labelling” drive took action against 3,704 accounts; and the Provisions on Simplified Personal Information Protection Measures for Small Personal Information Processors (Order No. 25) has been promulgated. Regulatory density — and its coupling with corporate reputation governance — has risen markedly.
Case Insight This Month: Louis Vuitton (LV) twice became the centre of trademark-related public opinion within half a month — the Suzhou Intermediate People’s Court ordered Molly Tea (Mo Li Nai Bai) at first instance to pay RMB 10.3 million in damages, and the Beijing Intellectual Property Court held a hearing in the trademark administrative case LV brought, for the sixth time, against CNIPA. In the same period, Driscoll’s faced a crisis of trust in China triggered by a US class action over PFAS in strawberries, and a safety controversy over ZARA’s wide-leg trousers spread from overseas into China. Multinational brands in China are simultaneously facing two categories of risk — product safety and trademark enforcement backlash — which cross-contaminate each other in the court of public opinion; the increasingly negative “foreign enforcement” narrative is eroding the discursive space for legitimate rights enforcement.
Counsel’s Recommendation This Month: Bring the analytical framework for each of the two scenarios forward into the legal decision-making stage. Before initiating trademark enforcement, first run the four questions — “good faith of the rights, profile of the counterparty, dissemination narrative, and timing window” — and only once these are cleared choose a litigation posture of either “litigating to settle” or “litigating to deter.” Once a product safety incident occurs, legal and PR should be launched in parallel: immediately verify the reporting, notification, and recall obligations under Article 19 of the Law of the People’s Republic of China on the Protection of Consumer Rights and Interests and the two-working-day reporting deadline under Article 8 of the Interim Provisions on the Administration of Consumer Product Recalls, with external messaging following in step after legal review — statutory actions must not be held back behind the PR timetable.

Governance Signals This Month

This month’s six governance signals advanced along three tracks — “dedicated legislation, special campaigns, and detailed rules” — both adding tools for the protection of corporate reputation and drawing new red lines for content generation and dissemination. The following sets out, item by item, the background, main content, and implications for corporate reputation governance.

I. The Cyberspace Administration of China Solicits Public Comments on the “Anti-Cyberviolence Law of the People’s Republic of China (Draft for Comments)”

Issuing Authority: Cyberspace Administration of China (CAC)

Date of Issue: Released July 29, 2026; comment period closes August 28, 2026

On July 29, 2026, the CAC, together with relevant departments, drafted the Anti-Cyberviolence Law of the People’s Republic of China (Draft for Comments) and released it for public comment, with a feedback deadline of August 28, 2026. This is China’s first dedicated legislation on cyber-violence. It aims to prevent, stop, and punish cyber-violence and to protect the lawful rights and interests of individuals and organisations, filling the legislative gap left by the rules on cyber-violence governance previously scattered across the Civil Code, the Cybersecurity Law, and the Public Security Administration Punishments Law.

Article 2 of the Draft defines “cyber-violence” by enumeration, including the concentrated publication of information containing insults and abuse, rumour-mongering and defamation, incitement of hatred, provocation of antagonism, intimidation and coercion, or discrimination and prejudice; the unlawful concentrated publication of others’ personal information; persistent online intimidation and online harassment; and other cyber-violence activities that infringe lawful rights and interests. This definition draws types of conduct previously dispersed across various branches of law into a single unified concept, and corresponds closely to scenarios commonly seen in corporate reputation governance, such as “organised attacks,” “doxxing,” and “persistent harassment.” Compared with the boundary of the public-opinion-oversight exemption set by Article 1025 of the Civil Code and the “notice-and-takedown” rule for online infringement established by Article 1195, this law makes “concentrated” and “persistent” constituent elements, in substance providing a clearer yardstick for distinguishing “legitimate criticism” from “cyber-violence.”

For enterprises, once this law takes effect, when an enterprise or its executives encounter concentrated, organised online attacks, they may invoke Article 2 to characterise the conduct, apply for a personality rights injunction under Article 997 of the Civil Code, and require platforms under Article 1195 to take necessary measures such as deletion, blocking, and link disconnection. Enterprises are advised to have their legal teams study the Draft and submit feedback before the August 28 deadline, focusing on the criteria for identifying “organised attacks” and the degree of detail in platforms’ handling obligations, and to embed “anti-cyber-violence” provisions into their online reputation contingency plans and executive protection mechanisms.

Original text: https://www.cac.gov.cn/2026-07/29/c_1787072711938509.htm

II. The Office of the Central Cyberspace Affairs Commission Carries Forward the First Phase of the “Qinglang · Rectifying AI Application Irregularities” Special Campaign

Issuing Authority: Office of the Central Cyberspace Affairs Commission

Date of Issue: July 6, 2026

On July 6, 2026, the Office of the Central Cyberspace Affairs Commission reported on progress in the first phase of the “Qinglang · Rectifying AI Application Irregularities” special campaign. Launched in April 2026, the campaign focuses on irregularities such as failure to fulfil large-model filing and registration obligations as required, inadequate safety and review-filtering capabilities of AI platforms, AI data poisoning, and inadequate implementation of labelling for generated and synthesised content, and has opened a dedicated “AI Application Irregularities Reporting Zone” to receive public reports.

In the first phase, more than 14,000 non-compliant AI products — including websites, apps, and AI agents — were dealt with; more than 6 million items of illegal and non-compliant information were removed; action was taken against more than 26,000 accounts; and more than 1,300 non-compliant AI products for sale and 9 non-compliant open-source datasets were taken down. Local cyberspace authorities adopted supervisory measures suited to local conditions: for example, the Beijing Cyberspace Administration established a coordinated mechanism of “platform self-inspection + routine patrols + technical screening + lead verification”; the Shanghai Cyberspace Administration guided platforms to upgrade their reporting mechanisms; and the Zhejiang Cyberspace Administration carried out rectification across multiple dimensions, including model review capability, training corpus security, and AI data poisoning. The Office of the Central Cyberspace Affairs Commission made clear that the next phase will focus on prominent problems such as using AI technology to produce and publish false information, impersonate or counterfeit others, infringe the rights and interests of minors, and engage in online “water army” (astroturfing) activities. In substance, this campaign translates the labelling obligation for generated and synthesised content under the Provisions on the Administration of Deep Synthesis of Internet-based Information Services and the filing and registration obligation under the Interim Measures for the Administration of Generative Artificial Intelligence Services into quantifiable enforcement data.

For enterprises using AI tools for marketing, customer service, and content generation, this campaign sends three signals: first, AI-generated content must be prominently labelled, or it may be deemed non-compliant information and removed; second, applications connected to unfiled large models face the risk of being taken down, so enterprises should verify the filing status of service providers before procuring AI services; third, accounts that use AI to generate negative enterprise-related information or to impersonate or counterfeit corporate executives have been made a focus of the next phase, and enterprises may on this basis file complaints with platforms and cite the outcomes of this campaign in demanding action.

Original text: https://www.cac.gov.cn/2026-07/06/c_1785081384384987.htm

III. The Office of the Central Cyberspace Affairs Commission Takes Strict Action Against a Batch of Accounts and Platforms with Non-compliant Short-video Content Labelling

Issuing Authority: Office of the Central Cyberspace Affairs Commission

Date of Issue: July 10, 2026

On July 10, 2026, the Office of the Central Cyberspace Affairs Commission reported strict action against a batch of short-video accounts that published fictional or staged content without labelling it in accordance with the rules. These are the concentrated enforcement results since the deployment of short-video content labelling work in May 2026, targeting typical irregularities such as “exploiting the image of persons with disabilities to elicit pity,” “fabricating tragic storylines about food-delivery riders,” “staging mother-in-law/daughter-in-law conflicts to fuel anxiety,” and “faking acts of heroism to win goodwill.”

Over the past two months, key websites and platforms have removed more than 241,000 short videos violating the labelling rules, taken action against more than 21,000 accounts, and corrected or added labels to mislabelled or unlabelled short videos more than 1.275 million times. The report lists typical cases in four categories, all pointing to the same core requirement — short videos publishing fictional or staged content must be prominently labelled “fictional/staged”; accounts that fail to label, or label ambiguously, will be muted or shut down. For certain small and medium-sized platforms that have not strictly implemented the short-video content labelling requirements, the relevant local cyberspace authorities have been instructed to summon them for regulatory talks and take action. This level of enforcement translates the principled requirement in the Provisions on the Governance of the Online Information Content Ecosystem that “false information shall not be disseminated” into enforceable labelling rules and quantifiable enforcement data.

For enterprises: when brands place storyline-based or staged marketing content on short-video platforms, they must ensure it is prominently labelled “advertisement” or “staged,” so as to avoid being caught up in the labelling crackdown and subjected to traffic throttling or takedown; when encountering short videos that smear the enterprise through fictional storylines, they may cite this report in requiring platforms to take action under the labelling rules and remove the relevant content; and enterprises are advised to review, in parallel, the labelling compliance of their official short-video accounts, with particular attention to whether review-type and comparison-type content fully discloses information sources and interest relationships.

Original text: https://www.cac.gov.cn/2026-07/10/c_1785427788435908.htm

IV. Cyberspace Authorities Crack Down on “Self-media” Failure to Properly Label Information Sources

Issuing Authority: Cyberspace Administration of China

Date of Issue: July 27, 2026

On July 27, 2026, cyberspace authorities reported that they have continued to supervise and guide websites and platforms in advancing information source labelling. Recent inspections found that some “self-media” accounts, when publishing content relating to domestic and international current affairs, failed to properly label information sources, confusing the public and misleading audiences. Cyberspace authorities stepped up inspection and enforcement, urging websites and platforms to take action, in accordance with law and user agreements, against 3,704 non-compliant accounts, and to correct mislabelled or unlabelled short videos more than 48,000 times.

The report lists three typical scenarios: first, directly reposting or simply aggregating international current-affairs information without labelling its source, leaving users unable to trace it; second, using expressions such as “circulating online,” “netizen revelations,” or “foreign media reports” to deliberately evade review and obscure the information source; third, using exaggerated expressions such as “thunderous strike” or “they’re fighting again” as titles and cover images, accompanied by archival footage of international events or patched-together and edited online images and videos, to interpret events in a one-sided or distorted manner. Cyberspace authorities emphasised that information source labelling is a basic precondition for ensuring that online information content is true, accurate, credible, and traceable, and required websites and platforms to establish long-term institutional mechanisms for standardised labelling, so that every item requiring a source label is labelled. Together with this month’s short-video labelling crackdown, this forms a coherent thread of “labelling governance,” translating the requirement in the Measures for the Administration of Internet Information Services that information be true and accurate into specific operational rules.

For enterprises: when self-media accounts disseminate untrue enterprise-related information using vague expressions such as “circulating online” or “according to revelations,” enterprises may cite this report in requiring platforms to take action under the information source labelling rules, and demand that the self-media party provide the original source of the information; when publishing corporate news or responses, enterprises should ensure their own official accounts label information sources properly, so as to avoid traffic throttling or platform action due to labelling defects; and enterprises are advised to establish a mechanism for “tracing the sources of enterprise-related information,” monitoring and filing complaints against accounts that disseminate negative enterprise-related information under vague sources.

Original text: https://www.cac.gov.cn/2026-07/27/c_1786897886145841.htm

V. The Cyberspace Administration of China, the Ministry of Industry and Information Technology, and the Ministry of Public Security Release the “Measures for the Administration of Internet Information Services (Revised Draft for Comments)” for a Second Round of Public Comment

Issuing Authority: Cyberspace Administration of China; Ministry of Industry and Information Technology (MIIT); Ministry of Public Security

Date of Issue: Released July 3, 2026; comment period closes August 2, 2026

On July 3, 2026, the Measures for the Administration of Internet Information Services (Revised Draft for Comments), drafted by the CAC together with MIIT and the Ministry of Public Security, was released for a second round of public comment, with a feedback deadline of August 2, 2026. The Measures revise the 2000 Measures for the Administration of Internet Information Services (State Council Order No. 292), with the aim of aligning with the amended Cybersecurity Law and other laws and promoting the healthy and orderly development of internet information services.

Article 1 of the Revised Draft specifies the Cybersecurity Law as its legislative basis, upgrading information service administration from the early “licensing–filing” framework into comprehensive regulation coordinated with cybersecurity, data security, and personal information protection. The Draft was previously released for public comment once; this second round follows revisions made in light of public feedback, reflecting the legislature’s prudent approach to adjusting the rules in the information services sector. The revision addresses core issues such as licensing conditions for information services, content management rules, and the boundaries of platform liability, and represents an update to a foundational administrative regulation in the field of internet information services.

For enterprises that are internet information service providers or that conduct business through information services, attention should focus on the Revised Draft’s adjustments to platform liability, content management, and licensing conditions for information services, with legal teams organised to study the Draft and submit feedback before the August 2 deadline; the rules for handling untrue enterprise-related information may change as the Revised Draft is finalised, so enterprises should update in parallel the wording and cited authorities used in their online reputation complaints; and enterprises are advised to track the revision of these Measures together with this month’s Anti-Cyberviolence Law and Provisions on Simplified Personal Information Protection Measures for Small Personal Information Processors, building an overall framework for corporate internet compliance.

Original text: https://www.cac.gov.cn/2026-07/03/c_1784822399677167.htm

VI. The Cyberspace Administration of China and the Ministry of Public Security Promulgate the “Provisions on Simplified Personal Information Protection Measures for Small Personal Information Processors”

Issuing Authority: Cyberspace Administration of China; Ministry of Public Security

Date of Issue: Promulgated July 24, 2026 (Order No. 25); effective September 1, 2026

On July 24, 2026, the CAC and the Ministry of Public Security jointly promulgated the Provisions on Simplified Personal Information Protection Measures for Small Personal Information Processors as Order No. 25, effective September 1, 2026. This is an important supporting regulation implementing the requirement of the Personal Information Protection Law that “the national cyberspace authority shall coordinate relevant departments in formulating special personal information protection rules for small personal information processors,” and aims to lower compliance costs for small processors while holding the bottom line of personal information security.

Article 2 of the Provisions specifies that a “small personal information processor” is a personal information processor that processes the personal information of fewer than 100,000 individuals; Article 3 supports such processors in adopting simplified measures commensurate with their scale and capabilities to safeguard personal information security. Article 4 simplifies the manner of disclosing personal information processing rules — where information is collected offline, the rules may be disclosed by posting a notice in a conspicuous location at the business premises; where collected online, by means of service agreements, pop-up windows, website notices, and the like. These provisions translate the principled requirements of the Personal Information Protection Law into enforceable operational rules, both easing the burden on small, medium, and micro enterprises and making clear the bottom line that “simplification” does not mean “exemption.”

Small, medium, and micro enterprises should first determine whether they meet the “small personal information processor” criterion (processing the personal information of fewer than 100,000 individuals), and on that basis complete the simplified disclosure of their personal information processing rules before September 1; even where simplified measures apply, they must still comply with the basic obligations under the Personal Information Protection Law and the Regulations on Network Data Security Management — for example, in the event of a data breach, notification and reporting are still required in accordance with law; and enterprises are advised to apply these Provisions together with last month’s Measures for Network Data Security Risk Assessment, building a complete closed loop of personal information protection covering “compliance documentation — simplified measures — emergency response.”

Original text: https://www.cac.gov.cn/2026-07/24/c_1786638889704872.htm

Reputation Enforcement Cases This Month

I. PFAS Detected in Driscoll’s Strawberries Triggers US Class Action: The Evidentiary Boundaries of a Cross-border Supply Chain Separation Claim

Factual Background: According to a Jiemian News report of July 16, 2026, six consumers residing respectively in New York, New Jersey, Massachusetts, and Illinois filed a class action on June 26, 2026 in the United States District Court for the Northern District of California, alleging that multiple PFAS were detected in fresh strawberries from berry supplier Driscoll’s (Source: Jiemian News). The same report states that the plaintiffs’ core evidence is testing commissioned by the consumer advocacy organisation Mamavation and conducted by a US EPA-certified laboratory — the conventional strawberry samples submitted were found to contain 12 pesticide residues, 8 of which are PFAS compounds, while the organic strawberry samples submitted in the same batch showed no such residues (Source: Jiemian News). PFAS — per- and polyfluoroalkyl substances — are known as “forever chemicals” because their stable structure makes them difficult to degrade; their representative substances, perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS), have been classified by the International Agency for Research on Cancer under the World Health Organization as a Group 1 carcinogen and a Group 2B possible carcinogen, respectively (Source: Jiemian News).

It should be noted that, according to Jiemian News, the complaint does not provide direct evidence that consumers suffered health damage from eating the strawberries in question; its claims rest on consumer fraud and false advertising, not personal injury (Source: Jiemian News). Earlier, on June 24, a former food safety compliance manager for Driscoll’s US and Canada region had filed a separate suit in the Ventura County Superior Court in California, alleging that warnings they had raised during their tenure regarding compliance risks in pesticide use on certain farms were not taken seriously and that they were subsequently dismissed; Driscoll’s has denied the allegations in both cases (Source: Jiemian News).

The explanations concerning the Chinese market all currently come from the company itself. According to a Chuanguan News report of July 21, 2026, Driscoll’s official customer service stated that day that the incident originated in the US market and that the Chinese market operates on a separate supply chain, that berries sold in China are produced at its Yunnan base, and that the company has a nationally certified CMA testing laboratory; customer service also provided a screenshot of an explanatory note recording that the General Manager of Driscoll’s China had told customers and partners on July 17 that all strawberries sold in the Chinese market come from the Yunnan growing region (Source: Chuanguan News). On the afternoon of the same day, Driscoll’s China issued its Statement of Driscoll’s China Regarding Recent Overseas Information, stating that some discussions on domestic online platforms had misinterpreted the nature, claims, and scope of application of the two pending US lawsuits, that products sold in the Chinese market are unrelated to the facts, parties, and legal disputes involved in those two cases, and that its products comply with national food safety standards, expressly citing the requirements on maximum pesticide residue limits in National Food Safety Standard — Maximum Residue Limits for Pesticides in Foods (GB 2763-2026) and the requirements on contaminant limits in National Food Safety Standard — Maximum Levels of Contaminants in Foods (GB 2762-2025) (Source: Chuanguan News). All of the above are unilateral statements by the company; as of this report’s information cut-off date of August 11, 2026, no public information was found of any authoritative third-party institution or regulatory authority having made a determination as to their truthfulness.

Outcome or Current Status: According to Jiemian News, both US lawsuits remain pending, and Driscoll’s denies all allegations (Source: Jiemian News). As for domestic channels, according to a Chuanguan News report of July 21, 2026, as of the time of that report, at high-end supermarkets such as Sam’s Club and Ole’ “Driscoll’s products on sale could no longer be found” (the report’s original wording, which does not specify whether this refers to online searches or offline channels), while Freshippo (Hema), Yonghui, and others showed “restocking,” and Driscoll’s official self-operated channels in China continued to sell normally (Source: Chuanguan News). As of this report’s information cut-off date of August 11, 2026, no public information was found of the above supermarket channels resuming listings, nor of any domestic regulatory authority publicly intervening in the incident.

Legal Issues: The issues centre on three points. The first is the triggering conditions for product liability. Article 46 of the Product Quality Law of the People’s Republic of China (2018 Amendment) defines a “defect” as an unreasonable danger in a product that endangers personal safety or the safety of others’ property, and, where national or industry standards exist for safeguarding human health and personal and property safety, as non-conformity with such standards; Article 13 requires that industrial products that may endanger human health and personal or property safety must conform to the corresponding mandatory standards. As applied to this incident, these provisions come into play only if batches containing the residues at issue actually entered the Chinese market — which is precisely why whether the separation claim holds determines whether liability exists. The second is the evidentiary structure of the separation claim. The company asserted in the aforementioned statement that products in the Chinese market are unrelated to the two US cases (Source: Chuanguan News); legally, this assertion is an exculpatory fact, and the burden of proof lies with the company: origin segregation requires traceable records of growing bases, harvest batches, and cold-chain flows, and test data must be issued by a qualified institution under currently effective standards and cover the indicators at issue — an in-house CMA laboratory resolves the question of the issuing entity’s qualifications, but not whether the sampled batches cover the products at issue. The third is the boundary of characterisation in clarifying enterprise-related information. The class action rests on consumer fraud, and the former employee’s case is in nature a labour dispute; in dissemination, the two cases have been stacked together with the “carcinogen” label. The company must clarify misinterpretations, yet must not, while litigation is pending, make assertions about the facts of the cases beyond its own knowledge — once a clarification is proven false, the matter shifts from an online reputation problem to a false statement problem.

Counsel’s Observations: When a cross-border brand’s in-China reputation is affected by a product safety crisis involving its overseas parent, separation is almost an obligatory move; what determines success or failure is whether the separation claim can be independently verified. Driscoll’s response rested on three pillars: origin segregation, in-house qualified testing, and a legal distinction regarding the nature of the cases. All three are complete as legal argument, but all are the company’s own account, and their release was concentrated at the peak of public pressure — according to a Chuanguan News report of July 21, 2026, the topic “Driscoll’s carcinogen” hit the trending list that day (Source: Chuanguan News). Self-generated explanations released at such a moment typically suffer a trust discount for being “self-certification”: the public will ask whether test reports can be made public, whether sampling covers batches on sale, and how Yunnan-grown and imported batches are distinguished at store level. There is also a tension between long-term marketing narratives and crisis separation — a premium imported positioning emphasises uniform global quality, whereas a separation statement emphasises the independence of the Chinese supply chain; when both narratives coexist, the separation itself becomes a new point of attack. Accordingly, before issuing a separation statement, the company should first reconcile it with its past promotional messaging and confirm that the two can coexist on the same factual basis.

Implications for Enterprises: Enterprises with overseas affiliates should turn “overseas risk — in-China separation” into a contingency plan rather than an improvised response. Specifically: build up, in advance and by growing region and batch, test reports from qualified third parties, so that at the time of an incident the separation claim can be supported by verifiable evidence rather than mere statements; review the consistency of messaging between brand positioning publicity and supply chain statements — wherever publicity uses expressions such as “uniform global standards” or “the same supply chain,” assess before any crisis how these conflict with a separation claim; establish a tracking mechanism for pending overseas litigation, distinguishing the nature of claims (consumer fraud, personal injury, labour dispute) and preparing corresponding Chinese-language clarification templates; and proactively cooperating with domestic regulators’ spot checks — substituting external determinations for self-certification — is the most cost-effective path to minimising the trust discount.

II. First-instance Award of RMB 10.3 Million in LV v. Molly Tea Trademark Infringement Case: Cross-class Protection of Well-known Trademarks and the Public Opinion Tension with “Guochao” Brands

Factual Background: According to an IPEconomy report of July 3, 2026, the Suzhou Intermediate People’s Court rendered a first-instance judgment in the trademark infringement case brought by Louis Vuitton Malletier against Shenzhen Molly Tea Catering Management Co., Ltd. (深圳市茉莉奶白餐饮管理有限公司) and Dongxia Beverage Shop, Wuzhong Economic Development Zone (吴中经济开发区东侠饮品店). The court found that the four-leaf flower device used by Molly Tea was substantially similar to 7 registered four-leaf flower device trademarks held by LV and infringed the exclusive rights to use those 7 registered trademarks; it ordered Molly Tea’s principal company to pay LV, within ten days after the judgment takes effect, RMB 10 million for economic losses and RMB 300,000 for reasonable enforcement expenses, totalling RMB 10.3 million, with the franchised store involved jointly and severally liable up to a limit of RMB 100,000; Molly Tea was also required to publish a statement on the home pages of its official website, official Weibo, WeChat official account, Xiaohongshu, Douyin, and other platforms to eliminate the effects of the infringement (Source: IPEconomy).

As to the status of trademark rights, according to IPEconomy, CNIPA’s trademark search information shows that Molly Tea’s relevant trademark applications for the four-leaf flower device currently have a status of “rejection review” or “invalid” (Source: IPEconomy). From 2023 to 2024, the Molly Tea company filed multiple applications with CNIPA to register the four-leaf flower device, all of which were rejected; knowing that registration had been blocked, it changed its brand logo in the second half of 2024 to a black geometric four-leaf flower pattern (Source: Xinkuaibao (New Express)).

Outcome or Current Status: After the first-instance judgment, Molly Tea founder Zhang Bocheng stated that the company would appeal (Sources: IPEconomy, New Yellow River Dayu Finance). As to rectification status, according to a report released by New Yellow River Dayu Finance on July 5, 2026, a reporter at a Molly Tea store in Shanghai on July 4 saw that the black four-leaf flower device remained visible on cups and packaging bags; staff at multiple stores said the packaging was existing stock and that any replacement would await notice from headquarters, while official customer service said there was currently no information about replacing materials (Source: New Yellow River Dayu Finance).

Legal Issues: The first issue is the scope of application of cross-class protection for well-known trademarks. Article 13, Paragraph 3 of the Trademark Law of the People’s Republic of China (2019 Amendment) provides that where a trademark applied for registration in respect of non-identical or dissimilar goods is a reproduction, imitation, or translation of another party’s well-known trademark already registered in China, misleads the public, and may thereby damage the interests of the registrant of that well-known trademark, it shall not be registered and its use shall be prohibited. The focus of applying this paragraph is the element of “misleading the public” — extending from luggage and bags to the tea-beverage context requires showing whether the relevant public would be misled as to a co-branding, licensing, or cooperative relationship, rather than merely comparing whether the devices are similar. The public’s intuitive objection that “a milk tea costing a dozen or so yuan would not lead anyone to think it is a luxury product” points precisely to the room for determination on this element. The second issue is the boundary of the fair use defence. Article 59, Paragraph 1 of the same Law provides that where a registered trademark contains the generic name, device, or model of the goods concerned, the holder of the exclusive right to use the registered trademark has no right to prohibit others from using them fairly; whether the four-leaf flower motif is a traditional decorative element that should remain in the public domain determines whether this defence can succeed. The first-instance court’s rejection of Molly Tea’s defence means the court found that the specific expression of the device at issue exceeded the scope of a generic device. The third issue is the effect of subjective fault on the amount of damages. The rejection of a trademark registration application is a clear warning as to the status of rights; continuing thereafter to deploy a similar mark at scale as a core visual element is highly likely, in infringement litigation, to be found to evidence an intent to free-ride, directly driving up the amount of damages.

Counsel’s Observations: What is distinctive about this case is that legal victory and public opinion backlash occurred simultaneously. According to a Xinkuaibao report of July 16, 2026, after LV launched one trademark lawsuit after another, comments such as “won the lawsuit, lost the public” appeared on social media; the public’s concern was that an international brand was using trademark rights to make a long-established floral motif its exclusive property, squeezing the design and business space available to small and medium-sized operators (Source: Xinkuaibao (New Express)). This is a reminder to enterprises: a solid legal basis for rights does not mean that enforcement actions are legitimate in the court of public opinion. When the target, manner, and timing of enforcement combine to create a public perception of “the strong bullying the weak” or “cultural appropriation,” the winning judgment itself becomes material for a negative narrative. And according to a report released by New Yellow River Dayu Finance on July 5, 2026, on July 4 orders were queuing at multiple stores in Shanghai and Guangzhou, with one Guangzhou store having a queue of 106 cups (Source: New Yellow River Dayu Finance) — at least in the first few days after the first-instance judgment, store business activity and the litigation loss were clearly decoupled. This snapshot suggests that once a strong-versus-weak narrative takes shape, litigation outcomes and goodwill gains or losses do not necessarily move in the same direction. From the rights holder’s perspective, the economic value of such a judgment may be substantially offset — or even outweighed — by reputational costs; from the defendant’s perspective, public sympathy is not a ground for exemption from liability, and continuing to use the mark at issue after losing at first instance keeps adding to the base on which infringement damages are calculated, regardless of whether the judgment is ultimately upheld.

Implications for Enterprises: For chain and franchise brands, the trademark risk of visual identifiers should be handled before the design is finalised, not after launch: complete clearance searches and a distinctiveness assessment before finalising the brand logo, with particular attention to whether it is similar to a prior well-known trademark; once a registration application is rejected, immediately activate an alternative rather than continuing to use the mark in the hope of getting away with it — the record of a blocked application will become direct evidence of subjective fault in later litigation. After being sued, loss mitigation and procedure should proceed in parallel: while an appeal preserves procedural room, rectification of online interfaces, in-store materials, and packaging consumables must be completed in step, as any lag in any one of these will be fixed as evidence of “continued use after the judgment.” Franchise systems should also note that in this case the store involved was held jointly and severally liable up to a limit; responsibility for replacing materials, cost-sharing, and breach clauses between headquarters and franchisees should be clarified together in the rectification plan.

Key Online Reputation Incident Review This Month

Incident Overview

In July 2026, a safety controversy over ZARA’s floor-length wide-leg trousers spread from overseas social platforms into China. According to a Nanchang Evening News report of July 21, 2026, large numbers of fashion bloggers and ordinary consumers overseas had recently spoken out together, alleging that a lightweight, floor-length style of ZARA wide-leg trousers had a design defect, with some wearers falling on level ground and even suffering fractures; “deadly ZARA wide-leg trousers” became a hot topic on European and American social platforms (Source: Nanchang Evening News). The same report states that a consumer living in Toronto, Canada, Irene Costa, said she fractured her wrist and injured her face after falling while wearing the trousers, and that her doctor told her she would be left with a permanent facial scar; influencer bloggers Jessica Pierce and Camila Ribera Roca each posted videos of themselves falling on a level pavement without any external interference, with Camila’s video exceeding 1.2 million views; and several other netizens commented that wearing this style of trousers had caused them to fall, resulting in kneecap fractures and soft-tissue contusions (Source: Nanchang Evening News). The risk mechanism identified by the various parties converges on a single point: extra-wide legs combined with floor-sweeping length, paired with lightweight polyester fabric, cause the hems to swing widely when walking, making them easy to step on and trip over.

The company-side response emerged through customer service channels. According to a Chinanews Economic (Zhongxin Jingwei) report of July 21, 2026, customer service at ZARA’s flagship store on Taobao said that day that the style in question was sold out, and as to whether it had been delisted, said the matter had been logged and passed to the relevant department and that a reply would follow after further verification; customer service at ZARA’s flagship store on Douyin said the trousers were 100% polyester, that some sizes were sold out, and that there was as yet no unified official statement; regarding the problem of hems being easily stepped on and causing falls, that customer service representative suggested that consumers “cut the trouser legs after purchasing the product” to reduce such incidents (Source: Chinanews Economic). According to a Nanchang Evening News report of the same day, as of its publication ZARA had made no public response to this series of safety incidents; a reporter searching ZARA’s official China website that day found more than ten similar-cut items on sale, including three-dimensional panelled floor-length wide-leg trousers and linen-blend wide-leg casual trousers, priced from 259 yuan to 499 yuan, all retaining the extra-wide, long floor-length design; the report did not mention any relevant risk warning on the pages (Source: Nanchang Evening News). As of this report’s information cut-off date of August 11, 2026, no information was found of the brand issuing a public statement or initiating a recall in connection with the controversy.

Online Reputation Flashpoints

According to Shiwei business intelligence monitoring, from 14:00 on July 20 to 14:00 on July 22, 2026, the online reputation heat index for “ZARA wide-leg trousers safety hazard” was approximately 16,444, peaking at 08:00 on July 21; by platform, social networks accounted for approximately 69.5%, news apps approximately 18.0%, and short video approximately 7.7%; by sentiment, negative accounted for approximately 44.0%, neutral approximately 32.3%, and positive approximately 23.7% (Source: Shiwei Technology). Judging from the above data and the timeline of media reports, this round of online reputation had three drivers: first, after overseas videos of falls were reposted to domestic platforms, the footage itself carried extremely strong intuitive persuasive force, conveying risk perception without any textual argument; second, on July 21 mainstream media followed up with concentrated interviews, elevating overseas consumers’ individual complaints into a domestic public safety issue; third, after customer service’s response “suggesting cutting the trouser legs oneself” spread through media reports, the focus of the controversy shifted from “whether the product is defective” to “whether the brand is shirking responsibility” — the latter being far more damaging than the former, because it requires no professional judgment, and ordinary consumers can reach a conclusion on common sense alone.

Legal Risk Points

The first risk point is the determination of a defect. China’s current apparel standards impose no specific restrictions on the tripping risk of this type of trouser cut, but this does not place the problem outside legal evaluation. Article 46 of the Product Quality Law of the People’s Republic of China (2018 Amendment) defines a “defect” as an unreasonable danger in a product that endangers personal safety or the safety of others’ property; Article 3 of the Interim Provisions on the Administration of Consumer Product Recalls goes further, defining a defect as an unreasonable danger to personal or property safety that commonly exists in consumer products of the same batch, model, or category due to design, manufacturing, warnings, or other causes — “design” and “warnings” are listed side by side as causes of defects, meaning that a tripping risk caused by the cut design, and the absence of warnings in itself, fall within the textual reach of the norms and cannot be excluded as a matter of course on the ground that “there is no corresponding apparel safety standard.”

The second risk point is the continuing gap in the duty to warn. Item (5) of Article 27 of the Product Quality Law requires that products which are easily damaged by improper use, or which may endanger personal or property safety, bear warning signs or Chinese-language warning statements; Article 18, Paragraph 1 of the Law of the People’s Republic of China on the Protection of Consumer Rights and Interests further requires that, for goods that may endanger personal or property safety, business operators provide truthful explanations and clear warnings and explain methods for preventing harm. Given that multiple accounts of fall injuries have already emerged overseas and similar cuts remain on sale on the official domestic website, each additional day the warning gap persists thickens the basis for a future finding of “continuing to sell despite knowing the risk.”

The third risk point is the reporting, notification, and recall obligations. Article 19 of the Consumer Rights and Interests Protection Law provides that where a business operator discovers that goods it provides are defective and pose a danger to personal or property safety, it shall immediately report to the relevant administrative department and notify consumers, and adopt measures such as suspending sales, issuing warnings, and recalling the goods; where a recall is adopted, the business operator shall bear the necessary expenses incurred by consumers as a result of the recall. Article 8 of the Interim Provisions on the Administration of Consumer Product Recalls provides that where a producer or other business operator discovers that consumer products it produces or sells have caused or may cause death, serious personal injury, or major property damage, or have been recalled outside China, it shall report to the provincial-level market regulation department where it is located within two working days of discovery; Article 9 requires a producer that discovers a consumer product may be defective to immediately organise an investigation and analysis. Of particular importance to multinational brands is Article 29 — for imported consumer products, the entity designated by the overseas producer to carry out recalls within China is deemed the producer; where none is designated, the importer is deemed the producer. This provision forecloses any inference that “production is overseas, so the recall obligation does not lie in China.”

The fourth risk point is the legal consequence of shifting remediation costs onto consumers. The suggestion to “cut the trouser legs oneself” falls squarely on the wrong side of Article 19 of the Consumer Rights and Interests Protection Law: that Article expressly provides that the necessary expenses of a recall are borne by the business operator, whereas the cutting suggestion asks consumers to modify the garment at their own expense to eliminate the risk. Article 20, Paragraph 2 of the same Law further requires business operators to give truthful and clear answers to consumers’ enquiries about the quality and method of use of goods; customer service scripts are therefore not merely a matter of service standards, but the interface through which a statutory obligation is performed. On the liability side, Item (7), Paragraph 1 of Article 56 of the same Law provides that refusing or delaying the adoption of measures such as suspending sales, issuing warnings, or recalling defective goods as ordered by the relevant administrative department may be subject to a fine — where there are no illegal gains, a fine of up to RMB 500,000 — and, in serious cases, an order to suspend business for rectification or revocation of the business licence; Article 55, Paragraph 2 provides that where a business operator knowingly provides defective goods, causing death or serious damage to the health of consumers, the victim is entitled to claim punitive damages of up to twice the losses suffered. As for damages in individual cases, Articles 41 and 43 of the Product Quality Law have already established the producer’s liability for compensation and the victim’s right to claim compensation from either the producer or the seller.

Response Assessment

The brand’s losses in this round of online reputation were concentrated in three areas, none of which is a question of communication technique. First, customer service messaging was released externally without legal review. “Cut them yourself” is a textbook example of converting a statutory remediation obligation into consumer self-help; legally it conflicts with Article 19 of the Consumer Rights and Interests Protection Law, and in terms of dissemination it provides a perfect target for attack — such expressions should have been intercepted in advance at the level of the customer service script library, rather than clarified after the online reputation issue had already escalated. Second, statutory actions were held back by the pace of public opinion. The combination of facts — multiple accounts of fall injuries overseas and similar cuts still on sale in China — is in itself sufficient to trigger the investigation and analysis obligation under Article 9 of the Interim Provisions on the Administration of Consumer Product Recalls; whether a defect is ultimately found is the conclusion of the investigation, not a precondition for launching it. If an enterprise conflates “investigation and analysis” with “admission of a defect,” it will choose silence precisely in the window when action is most needed. Third, the cross-border response was disjointed. The product was the same and the risk mechanism identical, yet the overseas controversy had persisted for weeks without any corresponding action in China — effectively leaving the timing of the response to be dictated by the pace of domestic media coverage. The direction for improvement is to institutionalise “home-country response — in-China synchronisation”: when a product safety incident involving the same product arises overseas, the China entity should immediately launch investigation and analysis and a warning assessment, and work backwards from the aforementioned two-working-day reporting deadline in designing internal procedures.

Implications for Enterprises

The handling of product safety online reputation incidents should launch legal and PR in parallel: verification and performance of statutory obligations start immediately, with external statements following in step after legal review. For apparel companies specifically, three starting points are available: first, incorporate cut-safety assessment into the pre-launch process, adding warning labels and usage instructions for styles that may cause tripping, entanglement, or snagging — this is both a requirement of Item (5) of Article 27 of the Product Quality Law and evidence for later asserting that reasonable care was exercised; second, the customer service script library must be subject to legal review, expressly prohibiting expressions of the “handle it yourself,” “at your own risk,” or “modify it yourself” type, with a mandatory channel for escalating safety-related enquiries to legal; third, establish a synchronisation mechanism for overseas product safety incidents involving the same product, specifying who is responsible for triggering an investigation, within what time a preliminary assessment must be completed, and in what circumstances reporting to authorities and notification of consumers are required — writing the answers to these three questions into institutional rules is worth more than preparing any statement after the fact.

In-Depth Theme This Month

Theme: Trademark Enforcement Backlash — A Methodology for “Legitimate Enforcement ≠ Online Reputation Safety” Drawn from LV’s Two Cases

In July 2026, LV twice became the centre of trademark-related public opinion within half a month. At the beginning of the month, the Suzhou Intermediate People’s Court ordered Molly Tea at first instance to pay RMB 10.3 million in damages; on July 14, a hearing notice thrust “LV sues CNIPA” into the public spotlight. According to a report by The Beijing News of July 15, 2026, the Beijing Intellectual Property Court was scheduled to hold a public hearing at 09:30 on July 16 in a trademark administrative dispute bearing case number (2026) Jing 73 Xing Chu No. 4727 ((2026)京73行初4727号), in which the plaintiff is Louis Vuitton Malletier, the natural person Huang Minyao participates as a third party, and the defendant is CNIPA; the notice was dated May 27, 2026 and was issued by the court to notify the third party to appear and participate in the proceedings (Source: The Beijing News). The same report states that the subject of the dispute is a “four-petal flower” device trademark in Class 18; LV contended that the trademark was similar to its classic pattern and that the counterparty’s filing of numerous similar trademark applications violated the principle of good faith, but CNIPA, upon examination, did not accept its contentions; according to published judgments, there had previously been at least 5 first-instance trademark administrative cases with Louis Vuitton Malletier as plaintiff and CNIPA as defendant, of which 3 were upheld by the court and in 2 the claims were dismissed (Source: The Beijing News). According to a Xinkuaibao report of July 16, 2026, the third party Huang Minyao is a clothing business operator in Shantou, Guangdong, who since 2023 has applied in his personal capacity for more than 100 trademarks, 8 of which were opposed by LV after passing preliminary examination or being successfully registered, with 7 of those oppositions succeeding (Source: Xinkuaibao (New Express)).

Placing the two cases side by side reveals an easily overlooked distinction: one is a civil infringement lawsuit, the other a trademark grant-and-affirmation administrative lawsuit; the former resolves damages and cessation of infringement, while the latter reviews the legality of an administrative ruling — the two differ in legal nature, party structure, and adjudicative effect. Yet in the court of public opinion, the two cases were compressed into a single narrative — “LV is suing everyone.” This is the first source of online reputation risk in rights enforcement: the public does not classify by legal relationship, only by the relative strength of the actors. Enterprises in periods of intensive enforcement especially need to be aware that actions independent of each other in law will automatically be linked in dissemination into a single posture.

The second source is public questioning of the boundaries of rights. Article 59, Paragraph 1 of the Trademark Law of the People’s Republic of China (2019 Amendment) provides that where a registered trademark contains the generic name, device, or model of the goods concerned, or directly indicates the quality, main raw materials, function, use, weight, quantity, or other characteristics of the goods, the holder of the exclusive right to use the registered trademark has no right to prohibit others from using them fairly. The public’s questioning of the four-leaf flower motif points legally to precisely this Article: when a trademark device overlaps heavily in visual elements with a long-established traditional decorative motif, is what the rights holder can prohibit the motif itself, or the specific expression formed through use? Article 13, Paragraph 3 of the same Law provides cross-class protection for well-known trademarks, but conditions it on “misleading the public, such that the interests of the registrant of the well-known trademark may be damaged” — the further the protection extends across classes, the heavier the burden of argument on this element. These two provisions form a pair in tension: cross-class protection pushes the right outward, while the fair use defence pulls it inward; which side of this zone of tension an enforcement action falls on directly determines whether the public sees it as rights enforcement or as a land grab.

The third source is the evaluation of the legitimacy of the manner in which rights are exercised, and on this level there are clear adjudicative rules to follow. The key holding of Guiding Case No. 82, (2014) Min Ti Zi No. 24 ((2014)民提字第24号), states: “Where a party violates the principle of good faith, damages the lawful rights and interests of others, disrupts the order of fair market competition, maliciously acquires and exercises trademark rights, and asserts infringement against others, the people’s court shall, on the ground that this constitutes abuse of rights, render a judgment not supporting its claims.” (Note: the text in quotation marks is the original wording of the key holding of that Guiding Case; “maliciously acquires and exercises trademark rights” is the Supreme People’s Court’s standard for the constituent elements of abuse of rights, and is not this article’s independent characterisation of any party.) The practical significance of this rule lies not in setting an additional threshold for enforcement, but in turning “whether the rights were acquired in good faith” and “whether the manner of exercise is proportionate” into matters that can be examined in court. What an enterprise should ask itself before enforcing is precisely how much the counterparty could make of these two questions.

From this, a reusable sequence of self-review can be distilled. Step one is a good-faith self-review of the rights: whether the chain of rights acquisition is clear, whether it involves privatising elements of the public domain, and whether it may fall within the reach of the fair use defence. Step two is anticipating public perception: the profile of the enforcement target (whether it is a small or medium-sized merchant, whether it carries the attributes of a local cultural symbol), the intensity of the enforcement means (whether a letter and negotiation came first), and the timing of enforcement (whether it resonates negatively with other enforcement actions) — the narrative frame produced by the combination of these three often determines the course of online reputation more than legal victory or defeat. Step three is ranking the means by reversibility: reversible means such as letters, negotiation, and rectification demands should take priority, with the most irreversible litigation action placed last — what is irreversible is not only the procedure, but also the way public opinion fixes the enterprise’s image.

Compared with last month’s “Yu Jian Xiao Mian” trademark enforcement incident, the two cases share the same structure: a large brand’s legitimate enforcement backfiring through a “strong bullying the weak” narrative. The only difference is that in the “Yu Jian Xiao Mian” incident, the brand under question was a domestic chain, whereas this month it is a multinational brand, with the added cultural dimension of “foreign capital monopolising traditional Chinese motifs,” making the backlash correspondingly stronger. This also shows that whether a strong-versus-weak narrative takes hold has nothing to do with the enterprise’s nationality, and depends only on the relative size of the parties and the public character of the subject matter of enforcement. If enterprises simplistically read last month’s lesson as “only foreign companies get criticised,” they will repeat the same mistake in their next enforcement action.

Counsel’s Prescription

This month, multinational brands in China simultaneously encountered two categories of risk: product safety crises and trademark enforcement backlash. Although they belong to different areas of law, their handling logic shares the same core: legal actions must be evaluated within the online reputation coordinate system, and reputational risk assessment must be brought forward into the legal decision-making stage, rather than added after the action has been taken. The following addresses the two scenarios — product safety online reputation and trademark enforcement online reputation — in turn.

I. Product Safety Online Reputation: Launching Legal and PR in Parallel

In product safety online reputation incidents, legal and PR are not a relay run in sequence, but two lines launched in parallel: when an incident is triggered, the legal side’s defect assessment and verification of reporting and recall obligations, and the PR side’s online reputation monitoring and messaging preparation, should start at the same time. Parallel does not mean equal — external statements must be issued only after legal review, and statutory actions (reporting, recall, warnings) must not be delayed to wait for the PR timetable or internal sign-offs. In terms of legal consequences, what determines the enterprise’s position is the timing of performance of statutory obligations, not the wording of statements. The time-gradient response framework set out in the Prescription section of our June 2026 issue for “legitimate enforcement met with backlash” will not be re-argued here; instead, this section applies it to this month’s ZARA wide-leg trousers controversy and the Driscoll’s PFAS litigation, explaining item by item what should be verified at each stage.

The first task is to switch the concept of “defect” from a PR context back to a legal context. Article 46 of the Product Quality Law of the People’s Republic of China (2018 Amendment) defines a defect as an unreasonable danger endangering personal safety or the safety of others’ property, while Article 3 of the Interim Provisions on the Administration of Consumer Product Recalls lists design, manufacturing, and warnings side by side as causes. This means two things: first, the absence of a corresponding mandatory standard does not mean the absence of a defect — unreasonable danger is an independent criterion; second, the absence of warnings can in itself constitute a cause of defect, and an enterprise cannot answer “why was there no risk warning” with “the product met factory standards.” The core facts of the ZARA case — similar cuts on sale, no risk warning, and multiple accounts of fall injuries already emerging overseas — fall squarely at the intersection of this set of norms.

The second task is to distinguish “investigation and analysis” from “admission of a defect.” Article 9 of the Interim Provisions on the Administration of Consumer Product Recalls requires a producer that discovers a consumer product may be defective to immediately organise an investigation and analysis, and Article 8 requires that where a product has caused or may cause death, serious personal injury, or major property damage, or has been recalled outside China, a report be made to the provincial-level market regulation department where the operator is located within two working days of discovery. Launching an investigation and reporting are, in law, performance of obligations, not admissions of liability; if an enterprise conflates the two, it will choose silence precisely in the window when action is most needed, and instead leave evidence for a later finding of “continuing to sell despite knowing the risk.” Cross-border brands must also note Article 29 of these Provisions: for imported consumer products, the in-China recall entity designated by the overseas producer is deemed the producer, and where none is designated, the importer is deemed the producer — the China entity cannot hold the recall obligation in abeyance on the ground that the production stage is overseas.

The red lines for external statements should be drawn on the statutory text; the following three messaging rules can be derived directly from the provisions. Article 19 of the Law of the People’s Republic of China on the Protection of Consumer Rights and Interests requires business operators, upon discovering a defect, to immediately report to the administrative department and notify consumers, and to adopt measures such as suspending sales, warnings, and recalls, with the necessary recall expenses borne by the business operator; Article 20, Paragraph 2 requires truthful and clear answers to consumers’ enquiries. From these, three mandatory messaging rules follow: do not suggest that consumers modify the product themselves to avoid the risk, because this runs counter to the rule that recall expenses are borne by the business operator; do not leave safety enquiries in abeyance for long periods with lines such as “logged and passed on” or “no unified statement yet,” because the duty to answer is an express statutory obligation; and do not, before an investigation is complete, draw absolute conclusions of the “there is no quality problem whatsoever” type, because once overturned by subsequent testing, the problem escalates from product liability to false statement. The liability side should likewise be made visible to business departments: Item (7), Paragraph 1 of Article 56 of the Consumer Rights and Interests Protection Law provides for fines, suspension of business for rectification, and even revocation of the business licence for refusing or delaying an ordered recall, and Article 55, Paragraph 2 provides for punitive damages of up to twice the losses where an operator knowingly provides defective goods causing death or serious damage to health.

The characterisation of external information must likewise be managed in step. The Driscoll’s incident provides a sample from the other side: what the enterprise faced was not an allegation of defect in its own product, but the inaccurate stacking, within China, of information about litigation involving an overseas affiliate. The key to handling such situations is to distinguish “clarification” from “assertion” — it is permissible to explain that the class action’s claims rest on consumer fraud rather than personal injury and that the former employee’s case is in nature a labour dispute, as this is an accurate restatement of public information; but it is inadvisable to make conclusive judgments about the facts of the cases while litigation is pending. For enterprise-related information that is genuinely fabricated or misleading, the enterprise may, under Article 1195 of the Civil Code of the People’s Republic of China, send platforms a notice accompanied by preliminary evidence, requiring necessary measures such as deletion, blocking, and link disconnection; where the disseminator has a competitive relationship with the enterprise or a profit motive, the enterprise may assess asserting commercial disparagement under Article 12 of the Anti-Unfair Competition Law of the People’s Republic of China (2025 Revision) and invoking the penalties under Article 28. It should be emphasised that the civil, administrative, and criminal pathways differ in their constituent elements and standards of proof and must not be characterised by skipping levels; whether conduct is suspected of constituting a crime is for the judicial authorities to determine, and enterprises should not on their own make public characterisations.

II. Trademark Enforcement Online Reputation: Pass the Self-review First, Then Choose the Litigation Posture

This issue’s In-Depth Theme has already set out a four-question framework for pre-enforcement self-review; this section does not repeat that argument, but takes the next step: once the self-review is passed, what posture should the litigation itself take, and how should external communications be handled around litigation milestones.

Enforcement postures fall broadly into two types, and which to choose depends on the objective. One is “litigating to settle,” characterised by quick filing and quick mediation and a low profile — prioritising causes of action with lower evidentiary burdens, filing in the court of the defendant’s domicile or one conducive to mediation, and setting settlement terms focused on cessation of infringement, deletion, and a written apology without seeking publicity; it suits scenarios where the aim is to end the infringement as quickly as possible and where there is an ongoing commercial relationship with the counterparty or high public opinion sensitivity. The other is “litigating to deter,” characterised by open pursuit coordinated with communications, choosing causes of action with higher damages ceilings for deterrent effect; it suits scenarios such as organised smear campaigns, free-riding use, and disparagement by competitors, where a clear signal needs to be sent to the market. The two postures differ markedly in cost, duration, and public exposure; mixing them usually yields neither speed nor deterrence.

Which to choose can be approached through four factors. First, strength of evidence: where there is notarised direct evidence or an admission by the counterparty, open pursuit is suitable; where there is only weak evidence such as screenshots, settlement should be pursued to reduce evidentiary risk. Second, expected damages and costs: in addition to legal and notarisation fees, account should be taken of preservation costs, enforcement difficulty, time costs, and the hidden losses from secondary dissemination. Third, the impact on commercial relationships: where the counterparty is an important customer, supplier, or channel partner, the commercial cost of public litigation usually exceeds the damages gained. Fourth, the anticipated public opinion effect — also the most easily overlooked — whether the filing itself will become news, and whether it will create an impression of “the big bullying the small.” This month’s Molly Tea case is, in terms of strength of evidence, a typical strong-evidence case: the factual chain of continuing to use a similar mark after the trademark application was rejected is clear, making it suitable for “litigating to deter”; but the counterparty is a “guochao” (Chinese heritage-trend) tea-beverage brand carrying the attributes of a local cultural symbol, and the anticipated public opinion effect was highly negative — the two factors pointing in opposite directions. Such conflicts are precisely where decision-making is hardest: strong evidence is not a sufficient reason for open pursuit, and must be combined with an assessment of the capacity to withstand public opinion.

Around litigation milestones, several boundaries on external communications must be observed. If a statement is needed at the filing stage, it should be limited to stating the facts and claims already submitted to the court, without predicting the outcome; after a judgment is rendered, first verify whether it has taken effect before deciding how to communicate — taking this month’s Molly Tea case as an example, the defendant publicly stated after the first-instance judgment that it would appeal, and since the judgment has not been published and whether it has taken effect cannot be verified, any external statement treating the first-instance outcome as settled may be deemed misleading. Before a judgment takes effect, do not make characterising evaluations of the direction of the case, do not use the media to apply pressure on the proceedings, and do not publish statements with intimidating overtones (such as declaring what price the counterparty will be made to pay) — in the counterparty’s hands, such statements will be turned around to support a claim of abuse of litigation rights. As for an effective judgment itself, the facts it records carry strong public credibility and make excellent material for reputation repair, but citations should be strictly confined to what the judgment has found, without expansive interpretation.

Finally, the administrative grant-and-affirmation track and the civil infringement track should be communicated separately. LV’s suit against CNIPA is a trademark grant-and-affirmation administrative lawsuit, which reviews the legality of an administrative ruling and differs from civil infringement litigation in nature, parties, and adjudicative effect. When pursuing both tracks simultaneously, an enterprise’s external explanations should address them separately and avoid conflating them — the outcome of this month’s public opinion compressing the two cases into a single narrative shows that if the enterprise itself does not draw the distinction, the public certainly will not draw it on the enterprise’s behalf. Internally, the sequencing of the two tracks should be coordinated: the outcome of the grant-and-affirmation proceedings directly affects the rights basis of the infringement litigation, and where that outcome is uncertain, the degree of publicity of the infringement litigation should be correspondingly restrained.

Governance Toolbox This Month

This Month’s Toolbox: Product Safety Online Reputation Response SOP for Multinational Brands and Trademark Enforcement Online Reputation Risk Assessment Checklist

I. Product Safety Online Reputation Response SOP for Multinational Brands

Apart from deadlines expressly set by statute (such as the two-working-day reporting deadline), the following steps are ranked by priority; steps that can proceed in parallel should do so, and no self-imposed clock deadlines are set.

II. Trademark Enforcement Online Reputation Risk Assessment Checklist

Good-Faith-of-Rights Question: Were the rights acquired in good faith? Could the case fall within the scope of the fair use defence under Article 59, Paragraph 1 of the Trademark Law of the People’s Republic of China (2019 Amendment)? Are there circumstances of “knowing pre-emptive registration” or “batch litigation to force paid settlements” of the kind warned against in Guiding Case No. 82?
Counterparty-Profile Question: Is the enforcement target a small or medium-sized merchant? Does it carry the attributes of a local cultural symbol? Is it susceptible to being cast in a “big bullying the small” frame?
Dissemination-Narrative Question: Could the enforcement method be interpreted as “cultural appropriation” or “excessive enforcement”? The further the cross-class protection claim extends, the heavier the burden of argument on the “misleading the public” element (Article 13, Paragraph 3 of the Trademark Law (2019 Amendment))
Timing-Window Question: Is the enterprise in an IPO quiet period, financing, major transaction, or other sensitive window? Does the action resonate negatively with other recent enforcement actions?
Choice of Enforcement Means: Have reversible means such as letters, negotiation, and rectification demands been exhausted? Is the most irreversible litigation action placed last?
Choice of Litigation Posture: Based on the four factors — strength of evidence, expected damages and costs, impact on commercial relationships, and anticipated public opinion effect — determine whether this case will adopt “litigating to settle” or “litigating to deter,” and set the cause of action, jurisdiction, and communications plan accordingly
Review of Milestone Communications: At the filing stage, state only the facts already submitted to the court; after judgment, verify whether it has taken effect before making external statements; do not predict outcomes, do not apply pressure through the media, and do not publish intimidating statements
Two-Track Distinction: Address trademark grant-and-affirmation administrative litigation and civil infringement litigation separately, without conflating them in external messaging
Contingency Plan and Co-Sign-Off: Has a response plan and unified messaging been prepared for the “worst-case online reputation scenario”? Have legal, PR, and business co-signed? Has the wording of external spokespersons (including representing counsel) been subjected to online reputation review?

Usage instructions: If any of the four questions yields an unsettling answer, suspend initiation or adjust the approach. This form is recommended for embedding in the enterprise enforcement project initiation process, used in conjunction with the “Product Safety Online Reputation Response SOP.”

Disclaimer

This article is intended solely for risk analysis purposes and does not render any determination of illegality or infringement against any party; the merits of individual cases shall be determined by effective judicial decisions.