Bad Faith Trademark in China: 2026 Defense Guide

Here is a number that should make every foreign brand owner uncomfortable: 49.87 million. That is how many trademarks are currently registered in China. Somewhere in that ocean of marks, pieces of your brand probably already exist -- registered by someone who has no intention of using them, but every intention of selling them back to you.

42 percent of the Forbes Global 2000 have been hit. Not might be hit. Have been hit. Over 200,000 malicious applications get blocked every six months -- and that is just what CNIPA catches.

At Logo China Pro, we see this every single week. A brand comes to us ready to enter the Chinese market, only to discover someone already owns their name. The conversation always goes the same way: "But we invented this brand. How can someone else own it?" The answer is simple and brutal -- in China, registration beats reputation. Every time.

But here is what has changed: China's revised Trademark Law, effective January 1, 2027, is the most aggressive crackdown on bad faith filing in the law's 40-year history. For the first time, bad faith filers face real monetary penalties. For the first time, CNIPA can clean up the register on its own. The game is shifting -- and if you act now, you can use these new rules to your advantage.

This guide is not theoretical. Every data point comes from official sources -- CNIPA, the Supreme People's Court, state media. Every case is real. Every strategy has been tested.

Source: CNIPA 2025 Annual Data | People's Daily IP Report | CNIPA Revised Trademark Law Full Text

The Scale of Bad Faith Registration in China

China processes roughly four times the trademark application volume of the USPTO every year. That is a lot of opportunity -- and a lot of opportunity for someone to grab your brand before you get the chance. Here is what the official data actually shows (and it is worse than most people think):

Read those numbers again. 200,000 malicious registrations blocked -- every six months. And that is only what CNIPA caught. The ones that slipped through? That is where foreign brands get burned. The average settlement for a squatted mark runs 1.5 to 5 million RMB. One in three foreign companies reports launch delays of six months or more. This is not a "maybe someday" risk. It is a Tuesday.

Here is the structural problem, and we need to be blunt about it: in China, registration -- not use -- determines ownership. As Logo China Pro documented in its guide to China's first-to-file trademark system, your brand reputation means nothing without that registration certificate. That is what makes bad faith registration so profitable. Someone registers your name for $400, and you end up paying $420,000 to get it back. That math is not a typo.

Source: People's Daily (July 2026) | CGTN (April 2026)

What the 2026 Trademark Law Changes

On June 26, 2026, the 14th NPC Standing Committee passed the most comprehensive revision of China's Trademark Law since 1983. The new law expands from 8 chapters and 73 articles to 9 chapters and 87 articles. It takes effect on January 1, 2027. For a broader overview, see Logo China Pro's 5 Things to Do Before Jan 1 guide.

Why does this matter? Because for the first time in 40 years, bad faith filers face real financial consequences. Not just "your application gets rejected." Real fines. Real penalties. Here is what changed -- and why each change matters:

Article 19 -- The New Objective Test for Bad Faith

This is the centerpiece of the reform. The old Article 4 stated that "bad-faith applications not for the purpose of use shall be rejected." The problem was that "bad faith" is subjective -- difficult to prove, easy to dispute.

The new Article 19 replaces this with a two-part objective test:

1. "Not for the purpose of use" -- the applicant has no genuine intent to use the mark in commerce
2. "Clearly exceeds normal business needs" -- the volume or scope of filings is disproportionate to the applicant's actual commercial operations

Both conditions must be met for refusal. This shift from subjective intent to objective measurability is significant. It means CNIPA examiners can now look at filing volume, class coverage, and business scope -- concrete data points -- rather than trying to read someone's mind.

Article 24 -- Lowered Threshold for Pre-Emptive Registration Claims

The old Article 32 required proof that a mark was registered "by improper means" to challenge pre-emptive registrations of prior-used marks. The new Article 24 replaces "improper means" with "deliberately" (intentionally). This lowers the evidentiary bar -- proving someone acted deliberately is easier than proving they used "improper means."

The protection scope also expands from "prior rights" to "prior legitimate interests," covering trade names, product names, packaging, and other commercial identifiers beyond just registered trademarks.

Article 54 -- First Administrative Penalties (This Is the Big One)

For the first time in Chinese trademark history, the law attaches explicit monetary penalties to bad faith filing behavior at the application stage. Previously, rejection was the only consequence -- which meant serial squatters had literally nothing to lose. Now, a bad faith filer who causes "adverse effects" faces a warning and a fine of up to RMB 100,000 (approximately $14,910). Not catastrophic, but it finally means filing a fraudulent trademark is not a risk-free lottery ticket. The conditions:

• Knowingly violates Articles 15 or 16(1) (unauthorized agent/representative filings)
• Violates Article 19 (not for use, exceeding normal needs)
• Intentionally violates Articles 21, 22, or 24 (well-known mark infringement, agent filings, pre-emptive registration)

Article 57 -- CNIPA Can Now Clean House Without Waiting for You

Under the old system, cancelling an unused trademark required a third party to file a non-use cancellation petition. That meant brands had to find the squatted marks, hire counsel, and file the paperwork themselves. The new Article 57 empowers CNIPA to proactively cancel trademarks that have become generic names or have no legitimate reason for three consecutive years of non-use -- no petition needed. This is a game-changer because it means the government is now willing to clean up its own registry. For foreign brands, this means some of the defensive work you previously had to do yourself may happen automatically.

For a detailed walkthrough of the non-use cancellation mechanism, see Logo China Pro's guide on non-use cancellation.

Source: CNIPA Official Full Text | CNIPA Revision Summary

7 Real Cases That Show How Bad Faith Disputes Play Out

Legal provisions are important, but cases are where you see the real consequences. These seven cases span different industries, outcomes, and strategies. Some brands won. Some lost. All of them paid dearly. If you only read one section of this guide, read this one.

Case 1: Schiff -- 13 Years, 45 Cases, Supreme Court Victory (March 2026)

This is the case that should be required reading for every foreign brand entering China. A squatter named Huo Jianglin filed 82 trademarks copying the SCHIFF brand across 17 unrelated classes in 2011. Reckitt Benckiser (now RB Health) fought for thirteen years through 45 separate proceedings before China's Supreme People's Court finally ruled in their favor in March 2026. Thirteen years. Let that number sit with you.

The turning point was not the opposition filings -- those were all initially dismissed because the goods were deemed dissimilar. The breakthrough came when Reckitt pivoted to non-use cancellation, which succeeded because the squatter could not prove genuine use. That created the foundation for the final invalidation victory based on absolute grounds (Article 4 and Article 44(1)).

The lesson from Logo China Pro: Schiff won, but at what cost? Thirteen years of legal fees, internal resources, and management attention -- all for a battle that could have been prevented with a handful of preemptive filings totaling roughly $4,000 in official fees. Prevention is not just cheaper. It is faster. It is certain. Schiff is the exception that proves the rule: most brands do not have the patience or the budget for a 13-year war.

The Supreme Court found that Huo's pattern -- mass filing across unrelated classes, copying well-known brands and scenic place names, repeatedly re-applying after cancellations, publicly offering marks for sale online -- constituted clear bad faith that disrupted the registration order and harmed public interest.

Lesson: Persistence through all appeal levels pays off. Non-use cancellation is the strategic weapon when relative grounds fail. Absolute grounds (Articles 4/19, 44(1)) are more effective than similarity-based arguments when goods are in different classes.

Source: Supreme People's Court IP Tribunal

Case 2: Biloban Beer -- Agency Held Jointly Liable (2025)

In a landmark case selected as one of the 2025 SPC Typical IP Cases, a trademark agency was held financially liable for assisting a client's bad faith registrations. The client (a beer industry competitor) commissioned the agency to register 10+ marks confusingly similar to the established "Blue Girl" brand. After multiple prior rejections and court rulings found the marks similar, the client continued filing and licensed marks to others for profit.

The court ruled that the agency, as a professional trademark firm in the same region, should have known about the reputation of the "Blue Girl" marks and constituted contributory infringement by continuing to provide services. Damages: the client paid RMB 500,000; the agency bore joint liability for RMB 100,000.

Our take: This is the case that should keep trademark agencies up at night. Under the new Article 67, agency liability is further strengthened -- fines up to RMB 200,000 plus potential suspension of business. The era of "the client told me to do it" as a defense is over. Agencies now have real skin in the game, and that changes the entire ecosystem.

Source: Supreme People's Court

Case 3: Michael Jordan -- Partial Win, Chinese Name Only (2016)

Qiaodan Sports built a $500 million empire with 5,700 retail stores using "QIAODAN" and the Chinese characters for "Jordan." The Supreme People's Court ruled in 2016 that Michael Jordan had prior name rights in the Chinese characters, but NOT in the pinyin "QIAODAN." Some marks were cancelled, but Qiaodan Sports continued operating with most of its marks. No damages were awarded.

Lesson: Registering only the English name is insufficient. You must also register the Chinese character name AND the pinyin equivalent. For more on this, see Logo China Pro's guide on trademark squatters in China.

Case 4: Tesla -- Won After 3 Years, $420,000 in Costs

A businessman named Zhan Baosheng registered the Chinese name for Tesla and "Tesla" in Classes 9 and 12 in 2006 -- six years before Tesla entered China. He demanded RMB 20 million (roughly three times Tesla's initial China marketing budget) to transfer the marks. Tesla refused, filed for invalidation, and won in Beijing court in August 2014 after three years of litigation costing approximately RMB 3 million.

Post-victory, Tesla registered its Chinese name in all 45 classes plus variants and defensive marks, now holding over 280 trademarks in China.

Our take: Tesla won -- but spent roughly $420,000 and three years doing it. For a company worth billions. Now imagine you're a smaller brand. Can you afford that fight? The real lesson: well-known status evidence can overcome an earlier filing date, but prevention -- registering before you enter -- costs a fraction of the cure.

Case 5: MUJI -- Lost Textile Category After 20+ Years

Beijing Miantian registered the Chinese name for MUJI in Class 24 (textiles) in 2000, before MUJI entered China in 2005. Over two decades later, Miantian still holds those rights with 200+ stores and 300 e-commerce stores generating billions in annual sales. In 2025, the Supreme People's Court upheld Miantian's Class 24 trademark.

Our take: This is the most painful case in the entire database. MUJI -- one of the most recognized brands on earth -- lost its textile category in China because a squatter registered the name five years before MUJI entered the market. Twenty years later, the squatter is still winning. This is not a theoretical risk. This is a brand that loses millions in revenue every year from a single class it never protected.

Case 6: New Balance -- Won, But Lost the Chinese Name

New Balance registered its English name but not its Chinese nickname. A squatter registered the Chinese version and built a business around it. Multiple lawsuits followed: a 2015 case resulted in RMB 98 million (approximately $16 million) in damages; a 2022 Shanghai IP Court case awarded RMB 25 million (approximately $3.7 million). Total legal costs were enormous, and the Chinese name remains contested.

Our take: New Balance technically "won" -- but look at the scoreboard. Two lawsuits, over $19 million in damages awarded, and they still don't fully control their Chinese name. Winning in court and winning in the market are two different things. The Chinese nickname must be registered proactively, before someone else claims it.

Case 7: Under Armour -- First Well-Known Trademark Recognition (2026)

In 2026, the Beijing IP Court recognized "UNDER ARMOUR" as a well-known trademark for the first time in Chinese administrative litigation. A Nantong furniture factory had registered "UNDER ARMOUR" and a Chinese translation in Classes 34 (cigarettes) and 32 (beer). Cross-class protection was granted across three classes. The court established bad faith based on the squatter's pattern of copying multiple famous brands.

Our take: This is the breakthrough case everyone has been waiting for. Under Armour got what most foreign brands dream of -- first-ever well-known trademark recognition in Chinese administrative litigation. Once that status is locked in, cross-class protection follows automatically. The path exists now. It just requires the evidence and the persistence to walk it.

CNIPA's 10 Detection Scenarios for Bad Faith

How does CNIPA identify a bad faith filing? If you are wondering whether a particular application might get flagged -- or you want to know what evidence CNIPA is already looking at -- here are the ten specific circumstances from the updated CNIPA Trademark Examination and Adjudication Guideline (March 2026) that trigger a "not filed for the purpose of use" finding. Think of this as the government's checklist. If your squatter hits any of these, CNIPA can move against them on its own:

1. Mass filing: Excessively large number of applications clearly exceeding normal business needs, with no genuine intent to use
2. Large-scale copying: Copying or imitating prior trademarks of multiple entities with reputation or strong distinctiveness
3. Repeated filing: Repeatedly applying for the same entity's particular trademark with reputation or distinctiveness
4. Mass filing of others' business identifiers: Trade names, e-commerce names, domain names, product names, packaging, distinctive advertising slogans, appearance designs
5. Mass filing of public cultural resources: Names of well-known figures, titles of well-known works or characters, artistic works
6. Mass filing of geographical names: Administrative division names, mountain or river names, scenic spot names, building names
7. Mass filing of non-distinctive signs: Generic names, industry terms, directly descriptive terms
8. Mass filing followed by large-scale transfer: Filing a large volume of trademarks and transferring to dispersed assignees
9. Commercial exploitation: Selling trademarks at scale, coercing prior users into business cooperation, demanding exorbitant fees
10. Other circumstances: As determined by CNIPA on a case-by-case basis

CNIPA also considers the applicant's total filing volume, designated classes, transaction history, industry and business status, prior findings of bad faith, and similarity with trademarks or identifiers that have reputation.

There is an important distinction in how CNIPA treats two types of bad faith. Type 1 is non-use bad faith -- mass filing and hoarding that disrupts the registration order and harms public interest (Article 19). Type 2 is squatting bad faith -- free-riding on specific famous brands and capitalizing on trends (Articles 21, 22, 24). If Type 2 behavior also involves mass filing, both types apply concurrently.

Source: CNIPA Official Q&A

20 Defense Strategies for Foreign Brands

Based on the cases, the new law provisions, and CNIPA's enforcement direction, here are 20 actionable strategies organized by urgency. Logo China Pro has grouped these into three tiers: what you should do this month, what you should do in the next quarter, and what you need to have in place before January 1, 2027. Do not skim -- each one addresses a real gap that has cost brands real money.

Immediate Actions (Do This Month -- No Excuses)

1. Conduct a comprehensive trademark portfolio audit. Pull up every China trademark you hold -- active registrations, dormant marks, defensive filings across classes. Verify that the registrant is your company, not a distributor or partner. Identify marks that are unused or overbroad relative to your actual business scope.
2. Set up automated trademark monitoring. With the opposition window shrinking to 2 months under Article 36, monthly checks are too slow. Switch to weekly or daily monitoring of CNIPA's trademark gazette. Pre-authorize opposition decisions for clear-cut cases (identical marks in the same class).
3. Organize your use evidence file. Collect every sales record, e-commerce listing, social media screenshot, distribution agreement, exhibition record, and advertising material that shows your trademark in use in China. The new Article 78 makes use evidence a threshold requirement for claiming damages in infringement cases.

Short-Term Actions (Next 3 Months -- Before Someone Else Files)

4. File non-use cancellations against blocking registrations. If a squatter holds a mark that has been unused for 3 consecutive years, anyone can file for cancellation -- no standing requirement. The burden of proof shifts to the squatter, who typically cannot prove genuine use. Non-use cancellation was the turning point in the Schiff case.
5. Oppose pending bad faith applications immediately. The 2-month window under the new law means speed is critical. Prepare template opposition documents with standard arguments and evidence formats ready to customize.
6. Use absolute grounds when relative grounds fail. The Schiff case proved that when similarity-based arguments (relative grounds) do not work because goods are in different classes, absolute grounds like Article 19 or Article 44(1) can succeed. Focus on the squatter's filing pattern, not just your mark similarity.
7. Combine non-use cancellation with invalidation. Cancel the squatter's mark first for non-use, then follow up with an invalidation action based on bad faith. This two-step approach was the strategy that ultimately succeeded in the Schiff case after years of failed oppositions.
8. Monitor high-risk entities. Track filings by your former distributors, agents, franchisees, and employees. Under the new Article 24, agent/representative unauthorized filings carry explicit penalties. Build evidence of the business relationship to support future claims.
9. Document the squatter's pattern of behavior. If the same entity has filed multiple marks copying different brands, this pattern evidence supports both CNIPA examination findings and court determinations of bad faith.

Long-Term Strategy (Before Market Entry or Before Jan 1, 2027 -- This Is Where You Win or Lose)

10. Register the trio: English name + Chinese character name + pinyin version. The Jordan, Tesla, and New Balance cases all demonstrate the same lesson -- if you do not own the Chinese version of your name, someone else will. Register all three simultaneously.
11. Audit all 45 classes -- especially high-risk subclasses. China's unique subclass system means that registering in Class 25 (clothing) does not protect shoes, shoe polish, or sports bags. Class 25 alone has 13 subclasses. Map your business to the correct subclasses and consider broader coverage for core brand marks.
12. File before you fly. Register trademarks in China 12-18 months before planned market entry. The first-to-file system rewards registration, not use. MUJI's 20-year loss in the textile class happened because the squatter registered before MUJI entered China.
13. Register Chinese transliterations proactively. Whatever name Chinese customers might use for your brand -- register it. If you do not choose the name, someone else will, and they will register it first.
14. Cover all 45 classes for core brand marks. If budget permits, file your primary brand mark across all classes. Tesla now holds 280+ trademarks in China across all 45 classes plus defensive variants. The cost of full-class registration is a fraction of the cost of litigation.
15. Build well-known trademark evidence continuously. Under the new Article 21, cross-class protection extends to unregistered well-known marks -- a significant expansion. But you still need to prove well-known status through media coverage, sales data, awards, and market presence. Under Armour's 2026 victory was possible because of years of documented brand recognition.
16. Prepare for the Article 19 objective test on your own filings. If your company has historically filed broad defensive portfolios, prepare documentation justifying the scope -- business plans, market research, expansion timelines. The new objective test targets filings disconnected from actual business needs.
17. Preserve evidence of prior use in China. This is critical for well-known trademark claims, Article 24 pre-emptive registration challenges, and non-use cancellation defenses. Every shipment record, invoice, and marketing material is potential evidence in a future dispute.
18. Establish relationships with reputable Chinese trademark agents. Under the new Article 67, agencies face financial penalties for assisting bad faith filings. Working with established, reputable agencies reduces your risk of inadvertently being associated with bad faith behavior.
19. Budget for enforcement. Trademark protection in China is not a one-time registration cost. Budget for ongoing monitoring, periodic oppositions, and occasional litigation. The cases above show that costs range from $15,000 for a straightforward opposition to $420,000+ for multi-year Supreme Court battles.

The Bottom Line: Why the 2026 Law Changes the Game

For four decades, people said the same thing about China's trademark system: "It's first-to-file, and bad faith is just the cost of doing business." That was always a lazy conclusion. The 2026 revision makes it definitively wrong.

Does this eliminate bad faith? No. No legal system can. But it fundamentally changes the math -- and that matters more than any single enforcement action:

• For squatters: Administrative penalties up to $14,910 per filing. Agency liability up to $29,820. Credit discipline tracking. Proactive cancellation by CNIPA. The days of risk-free squatting are over. The cost of gambling just went up.
• For brand owners: Shorter opposition window (2 months) means you must act faster. But the tools -- objective bad faith tests, proactive cancellation, agency liability, broader prior interest protection -- are more powerful than anything that existed before. The question is not whether you have options. It is whether you use them in time.
• For the ecosystem: The combination of Article 19's objective test and Article 57's proactive cancellation means CNIPA can now clean up the register on its own initiative. This was never possible before. The government is no longer waiting for complaints.

The window to prepare closes on January 1, 2027. After that date, the new rules apply to every pending application, every opposition, and every enforcement action. Brands that have audited their portfolios, organized their evidence, and positioned their registrations will be protected. Brands that wait will be reacting -- and reacting under a tighter timeline, with less margin for error.

This is not alarmism. This is arithmetic. The brands that prepare now will spend thousands. The brands that wait will spend hundreds of thousands. Choose which side of that equation you want to be on.

FAQ: Bad Faith Trademark Registration in China

What is bad faith trademark registration in China?

Simply put: someone registers your brand name in China -- not because they want to build a business with it, but because they want to sell it back to you. Under the 2026 revised Trademark Law (effective January 1, 2027), Article 19 now establishes a clear objective test: if an application is not filed for genuine use and clearly exceeds normal business needs, it shall be refused. No more hiding behind "I intended to use it someday."

How many bad faith trademarks does China reject each year?

The scale is staggering. CNIPA data shows over 200,000 malicious registrations blocked per half-year period in both 2023 and 2024. Since 2023, 1.273 million deceptive or misleading trademark applications have been rejected cumulatively. And foreign companies filed over 8,400 opposition cases against bad faith registrations in 2025 alone. Source: People's Daily

What penalties does the 2026 Trademark Law impose on bad faith filers?

For the first time, real money is on the line. Article 54 introduces administrative fines of up to RMB 100,000 (approximately $14,910) for bad faith filings that cause adverse effects. Trademark agencies assisting bad faith filings face fines up to RMB 200,000 under Article 67. It is not a fortune -- but for the first time, the risk is not zero.

How long does it take to oppose a bad faith trademark?

The opposition window has shrunk to 2 months under the revised law (Article 36). The full process from filing to decision typically takes about 11 months. The new Article 57 also gives CNIPA proactive cancellation authority for unused marks -- meaning you may not have to file anything at all. CNIPA can act on its own.

What should foreign brands do first?

Two things, immediately. First, pull up every China trademark you hold and check for gaps -- wrong registrant name, missing classes, missing Chinese character versions. Second, set up automated monitoring so you catch conflicting applications within that 2-month window, not six months later. These two steps solve the two most expensive mistakes we see brands make.

This article is for informational purposes only and does not constitute legal advice. Trademark law is complex and fact-specific. For advice on your particular situation, consult a qualified trademark professional. Logo China Pro provides trademark filing coordination services, not legal representation.