chevron-down Created with Sketch Beta.

Doctored Trademark Specimens at the USPTO: Analysis of the Plague of Fake Specimens Threatening to Undermine the Principal Register

Eric J. Perrott

©2018. Published in Landslide, Vol. 11, No. 1, September/October 2018, by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association or the copyright holder.

The United States Patent and Trademark Office’s (USPTO’s) corps of trademark examining attorneys is grappling with a plague of fake, doctored, and digitally altered specimens, and there is little that the USPTO can do to prevent them. These fraudulent specimens threaten to bring the examination process to a grinding halt and undermine the integrity of the U.S. trademark application system; but worse, they could have potentially huge impacts on the future decisions of the USPTO leadership to raise fees and try to find fair ways to fight an issue that appears to consist mostly of foreign trademark applicants.

The USPTO and its staff are not the only ones affected. Private practitioners are already seeing the effects of the doctored specimens, from difficulties in the clearance process to increased scrutiny in the form of specimen refusals. Trademark attorneys, in-house counsel, and anyone affected by the U.S. trademark registration system should be aware of these issues, know what to look for, and know how to protect themselves and their clients from potential delays, clearance problems, and other new challenges in light of these fraudulent specimens.

Trademark Specimens: Proof of Use in Commerce

The U.S. trademark system is based on “use in commerce.”1 Without commercial use in the United States, a company or individual does not have protectable trademark rights. Pursuant to that basic pillar of trademark law, the U.S. trademark registration system, which gives the owners of trademark registrations certain invaluable benefits, requires proof of use in commerce for a vast majority of trademarks before a registration will issue.2 According to the USPTO Trademark Electronic Search System (TESS) database, of the almost 255,000 registrations issued in 2017, over 227,000 (89 percent) were registered under the basis of “use in commerce” in the United States, section 1(a) of the Lanham Act.

This means that in 2017, over 227,000 “proof of use” submissions were submitted, reviewed, and approved by the USPTO’s trademark examining attorneys and trademark specialists in the form of “trademark specimens.” The USPTO managed this workload and still kept ahead of its pendency goals and virtually all other metrics the USPTO uses to measure the productivity of the examining procedure.3

When reviewing an application filed under a “use in commerce” basis, trademark examining attorneys do not typically require invoices, bank statements, or any other actual proof that a transaction has taken place in order to show use in commerce. Instead, the Lanham Act and the trademark rules require trademark applicants to submit a sworn statement that their trademark is in use on all goods and services listed in the application, and file a “specimen showing how the applicant uses the mark in commerce.”4

Trademark specimens come in all shapes and sizes, and the acceptability of any given specimen depends on the individual goods and services. In general, a specimen for goods should consist of a label, tag, or container for the goods, or a display associated with the goods.5 For services, an applicant’s specimen must show the mark as used by the applicant to advertise or render the applicant’s services. According to USPTO guidance, this may include websites, advertisements, brochures, billboards, menus (for restaurants), and much more.6

Examining attorneys who receive a specimen that they believe does not meet the minimum standards of the trademark rules must issue an office action requiring a replacement, and in some cases requiring the applicant to submit additional information. Trademark applicants then have six months to respond to the examining attorney’s office action with new specimens or arguments against the initial rejection.

Issues with specimens are not new. For many attorneys, a part of the client counseling that comes with prosecuting a trademark application is advice on usage of the mark that meets the standards of the USPTO. With new technology and new delivery methods, the list of what the USPTO may accept as a specimen is always changing and evolving. However, this system, while subject to the individual discretion of the examining attorney assigned to the application, typically results in minimal delays before resolving the issues.

However, this system is built on the assumption that most trademark filers are filing in good faith and are not attempting to deceive the USPTO into granting a trademark registration by submitting a false, doctored specimen. With the influx in doctored specimens filed in bad faith, the trademark registration system will be tested in a way it has not been tested before.

Fraudulent Specimens: The Issue

Over the past year, there has been an unprecedented increase in the number of United States trademark applications with international applicants, and with them an epidemic of specimens that are seemingly legitimate on their face but turn out to be fraudulent. While specimen issues have been a problem for many different types of trademark applications in the past, the recent influx of applications with fraudulent specimens share several similarities, from their country of origin down to the format of their trademark applications.7

Filing Basis: TEAS Plus

The recent applications are generally filed as Trademark Electronic Application System (TEAS) Plus applications, which allows for the least expensive filing fee of $225 per international class. Each application usually includes a “translation statement” that the mark does not have a translation in a foreign language. This translation statement is important because, under the rules regarding the reduced-fee TEAS Plus filing, applicants forfeit this special lower-cost filing status if additional information, such as a translation statement, needs to be submitted.

Marks and Identifications: Arbitrary and Broad

The marks themselves vary, but typically they consist of random letters mixed together in an arbitrary way, or other arbitrary words mixed together. This is a particularly clever strategy because, by submitting this type of arbitrary mark, the applicants reduce the chance of receiving an initial refusal based on a likelihood of confusion with preexisting registrations and reduce the chances of a third party filing a notice of opposition against the application. This filing strategy greatly increases the chances of a successful registration.

Corporate Status

While there are applicants from all over the world who have specimen issues, the vast majority of the recent wave of filings are made by applicants from China. The applications are filed without counsel and are owned by foreign limited companies. Also, the applications are typically filed within hours of each other.

Doctored Specimens

The final similarity that all of these applications share is highly convincing doctored specimens. These range from stock images doctored to include a fake label to previously submitted specimens doctored to include a different trademark on the packaging or tags.

Why This Is a Problem for the USPTO and for Brand Owners

On its face, the increase of doctored specimens would not seem to be overly burdensome or problematic. However, trademark registration brings with it many valuable benefits and “confers important legal rights and benefits on trademark owners who register their marks.”8 As a result, the USPTO has instituted several programs to ensure the integrity of U.S. trademark registrations (the Principal Register). When a pilot study showed over 50 percent of renewals included items that were not actually still in use, the USPTO instituted a random audit program for renewals.9 When it was clear that signatories of the trademark applications were signing the sworn declarations without reading the requirements, the USPTO increased the solemnity of the declarations and included mandatory checkboxes to raise attention to these important declarations.10

When a registration is issued to a company or individual who does not have actual use of the trademark in the United States, that registration is given the same presumptions and benefits as a registered trademark with bona fide use. These fraudulent registrations that issue from applications using doctored specimens undermine the integrity of the Principal Register and ultimately weaken the importance of obtaining a registration.

In addition, the USPTO has set certain goals to measure its effectiveness. These goals measure both timeliness and quality, and USPTO examining attorneys are given a bonus based on helping reach these goals. Trademark applications are already increasing from last year and are at an all-time high. Now, examining attorneys must spend additional time scrutinizing each specimen to ensure it is not fake, or risk approval of a fraudulent specimen that provides the basis for a registration that never should have been issued in the first place.

As the USPTO attempts to hire new trademark examining attorneys to meet the increased number of U.S. trademark applications, it faces several unknowns. To keep the application fees low, the USPTO’s Trademark Division relies on renewal filings to meet budgetary goals.11 If these applicants are fraudulent, it is highly unlikely that they will file renewals five to six years later. This makes workforce planning for the USPTO extremely difficult and may lead to further turmoil by overstaffing in the future.

All of this has similar impacts on practitioners and trademark owners. When conducting a comprehensive trademark search, the first place a search professional will look is the USPTO’s Principal Register. Fraudulent registrations make it harder, more time consuming, more expensive, and riskier for a company to develop, select, and clear a trademark for a new product or service.

Trademark owners are also required to diligently protect their trademarks. Any action that is required by trademark owners to protect their marks against fraudulent trademark applications is a burden on those trademark owners to monitor, oppose, and otherwise enforce their rights against a potential infringer.

Finally, as a result of these fraudulent specimens, trademark examining attorneys and support staff are taking a stricter view on specimens and are more likely to issue specimen refusals, even when the specimens at issue may be genuine. The examining attorneys simply do not have the time or resources to closely check every specimen, so many trademark owners with legitimate specimens will get caught in the wide net cast by examining attorneys as they work to combat the fraudulent filings.

Why an Increase in Fraudulent Specimens?

Foreign Subsidies

“Why are foreign applicants making these filings?” The golden question. Since at least as early as 2012, USPTO officials have publicly expressed concerns over the Chinese government offering subsidies to Chinese companies and individuals who make foreign intellectual property filings. At the time, the focus was on patent filings made in the United States, subsidized by the Chinese government.

However, recent reports have shown that those subsidies extend to trademark filings as well. In 2016, under article XVI:1 of the General Agreement on Tariffs and Trade (GATT) and article 25 of the Agreement on Subsidies and Countervailing Measures (SCM Agreement), the Chinese representatives to the World Trade Organization (WTO) provided a list of subsidies offered by 20 of its 32 provinces. While many of the policies and subsidy programs have expired or been replaced with other programs, there is evidence of clear incentives for Chinese entities to register their brands abroad.

The amount of the subsidy was dependent on the number of countries in which trademark registration was achieved. When the United States questioned whether these programs violated article 3.1(a) of the SCM Agreement, the Chinese responded by stating that each of the three programs were no longer in effect. According to the WTO documents, on average, applicants who filed in the United States were entitled to receive RMB 5,000 (approximately USD 790) under the various programs.12

The method that the applicants use when filing these applications seems to be designed to obtain a registration as cheaply and with as little interaction with the USPTO as possible. As noted above, these are TEAS Plus applications with the lowest fees, and the applicants rarely, if ever, hire U.S. counsel to file the applications. One line of reasoning is that these applicants are defrauding both the Chinese subsidy program and the USPTO and, as a result, netting approximately $500 per successful trademark registration. Presumably, these applicants are also making filings in any other jurisdiction that can be filed pro se and with minimal cost.

Another line of reasoning is more disturbing. If the Chinese government is offering these incentives, are government officials actively encouraging these kinds of filings? If they are, then the sheer volume of these filings could be a coordinated, state-sponsored attack on the U.S. trademark registration system and the USPTO. So far, there has been no public evidence of this type of collusion between the Chinese government and the applicants, aside from the known subsidy programs for foreign trademark registrations.

As the filings continue to increase, the U.S. government will need to closely monitor the situation to ensure that the United States’ intellectual property system is not being manipulated by a foreign power.

Amazon Brand Registry and the Stress on Brands to Own Registrations as Quickly as Possible

There are other contributing factors to the rise of fraudulent specimens. In particular, one of the biggest online marketplaces in the word,, recently launched a new “brand registry” program that requires a participant to have a registered trademark to enroll. Suddenly, Amazon sellers from around the world have raced to register trademarks as quickly as possible for various brands being sold on the Amazon platform. The quickest way to register a trademark is under a “use in commerce” basis, filed with a specimen. This may also be contributing to the increase in fraudulent specimens.

Beyond Amazon, other online marketplaces and social media platforms are increasingly scrutinizing the intellectual property claims made against their sellers and users and often take faster action if a claim involves a registered trademark.

Identifying Fake Specimens: The TM Specimen Protests Email Pilot Program

In response to the increase in doctored specimens, on March 6, 2018, the USPTO announced the “TM Specimen Protests Email Pilot Program” (Pilot Program), a program to allow the public to submit notices to the government alerting it to potential fraudulent specimens that have not yet registered. Described as a “streamlined protest procedure for reporting improper specimens,” the Pilot Program provides a procedure for the public to follow to report potentially fraudulent specimens.

To submit a protest against a doctored specimen, the individual must send an e-mail to [email protected] with evidence of third-party use of an identical image, or prior registrations or applications that use the same image in connection with a different mark. Timing is important for this Pilot Program, because it is limited to section 1(a) “use based” applications that have not yet cleared the “publication period,” the 30-day period wherein third parties may oppose the registration of an application. All protest e-mails must be received no later than the thirtieth day after publication for opposition. Currently, the Pilot Program does not cover “statement of use” filings or section 8 renewal specimens.

Several questions remain about the scope and effectiveness of this Pilot Program. First, the USPTO has indicated that although it has no plans to make these “protests” public by posting them anywhere, the policy could change at any time. Furthermore, presumably all submissions would be subject to a Freedom of Information Act request or may be obtained for litigation reasons.

The effectiveness of a program like this remains to be seen in combating fraudulent filings, but it is another tool at the disposal of the USPTO as it continues to further its mission of maintaining the integrity of the trademark registration symbol. It also places a spotlight on all applications and sends a message to all applicants that submission of doctored or fake specimens will not be tolerated and could result in significant delays in obtaining (or even denial of) a trademark registration.

How to Identify Doctored Specimens

For practitioners and trademark owners, it is important to be able to identify doctored specimens. Some doctored specimens are simple to identify—clear evidence of manipulation is often present on a specimen that has been digitally altered, and when identified, the examining attorney issues an office action on those grounds. However, as the USPTO gets more effective at identifying the trends and signs that a specimen has been falsified, the applicants are changing tactics and producing doctored specimens that are increasingly more difficult to identify as “fake.”

The methods for detecting fraudulent specimens highlight the crux of the USPTO’s problems: photo-manipulation solutions are expensive and time-consuming, and require specialized skills in order to consistently catch. One method is through a reverse image search, such as Google Image Search or TinEye. These services compare the specimen against millions of images online and return likely matches. This method is effective in identifying doctored images based on stock images, but is less effective against images that are doctored images of prior specimens. Other websites use computer algorithms to view metadata and signs of manipulation automatically, including Izitru, ImageEdited, and PixelPeeper.

However, while computer algorithms may help save time, manual analysis of the photos for lighting anomalies, artifacts from resizing and editing, and other forensic techniques are still needed to detect fakes in many circumstances. Those techniques, mixed with knowledge of preexisting fraudulent specimens, can help effectively identify doctored specimens.

It is important for trademark attorneys, particularly those with foreign clients, to be able to recognize doctored images in order to provide clients with advice about the requirement for trademark use in the United States.

Fighting a Fraudulent Specimen

It is inevitable that trademark applicants and brand owners will be faced with a trademark application or registration containing a doctored specimen that is cited as a 2(d) refusal or as a prior pending mark that will likely be cited as a 2(d) refusal once registered. In cases such as this, the Pilot Program is unlikely to be helpful, because there are no guarantees that the USPTO will take action and the program is limited in timing and scope. Further, a potential adverse party may open itself up to adverse claims by submitting a report to the U.S. government in the event that the specimen was not doctored and use could be substantiated.

Notice of Opposition and Petition to Cancel

The first step for anyone affected by a potentially doctored specimen is to carefully analyze the specimens using the techniques mentioned above and consider hiring an expert to examine the specimens for authenticity. Further, applicants should perform a thorough investigation into whether the cited mark is in use on all goods and services listed in the application. Once the applicant has done its due diligence and found actual offers for sale and/or a doctored specimen, the applicant likely has a few different claims before the Trademark Trial and Appeal Board (TTAB) in an opposition or a cancellation proceeding.


Another claim available is a nonuse claim, which is helpful for applications filed under section 1(a). If a trademark was not used in commerce in the United States prior to the filing of the use-based trademark application, that application is void ab initio and can be opposed or canceled.13


Since the TTAB’s decision in In re Bose raised the fraud standard to “clear and convincing evidence” of intent to deceive the USPTO, fraud claims have been extremely difficult to prove.14 However, “even under the demanding standard set by Bose, the submission of a ‘doctored-up specimen’ to the PTO, if proven, could constitute evidence of deceptive intent.”15


The integrity of the Principal Register is one of the USPTO’s most important missions. Fraudulent trademark registrations undermine the value of a trademark registration and ultimately hurt legitimate businesses, which must clear and register their own trademarks in a sea of potentially fraudulent applications. The USPTO faces a daunting task in identifying and challenging applications filed in bad faith, while still keeping up with the day-to-day increase in trademark applications filed by legitimate businesses. There is no “silver bullet,” and, while the USPTO scrambles to find a solution, both trademark applicants and practitioners should expect to see changes in policy and increased scrutiny in their filings.


1. See 15 U.S.C. §§ 1051 et seq. Congress’s power to protect trademarks comes from the commerce clause, allowing Congress to “regulate Commerce . . . among the several States.” U.S. Const. art. I, § 8, cl. 3.

2. Trademark applications that eventually register through the extension of protection under the Madrid Protocol or by using a foreign registration as a basis under section 44(e) of the Lanham Act are not required to submit proof of commercial use to obtain a registration. However, they are unable to bring a claim under the Lanham Act for trademark infringement without established use and must still submit proof of use pursuant to the five- to six-year section 8 renewal.

3. See Data Visualization Center: Trademarks Dashboard, USPTO, (last visited Aug. 20, 2018).

4. 15 U.S.C. § 1051; 37 C.F.R. § 2.34(a)(iv).

5. 37 C.F.R. § 2.56(b)(1).

6. Trademark Manual of Examining Procedure (TMEP) § 1301.04 (Oct. 2017).

7. The following observations are made based on a small sample size of applications where the USPTO has issued a “doctored specimen” office action refusal.

8. B&B Hardware, Inc. v. Hargis Indus., Inc., 135 S. Ct. 1293, 1317 (2015).

9. Post Registration Proof of Use Audit Program, USPTO, (last visited Aug. 20, 2018).

10. USPTO, Trademark Public Advisory Committee Annual Report 2016, at 12 (2016),

11. See Fee Setting and Adjusting, USPTO, (follow “USPTO Fee Setting - Activity Based Information and Trademark Fee Unit Expense Methodology” hyperlink under “Earlier Fee Setting Information: Trademark fee setting effort with January 14, 2017 effective date”) (last visited Aug. 20, 2018).

12. Committee on Subsidies and Countervailing Measures, New and Full Notification Pursuant to Article XVI:1 of the GATT 1994 and Article 25 of the Agreement on Subsidies and Countervailing Measures: China, WTO Doc. G/SCM/N/95/CHN/Suppl.1, at 57, 58, 60 (July 29, 2016).

13. See CPC Int’l Inc. v. Skippy, Inc., 1987 WL 124285 (T.T.A.B. May 28, 1987) (holding that registration was void ab initio because the specimens did not demonstrate use in commerce). However, nonuse claims cannot be made against a registration on the Principal Register registered for more than five years.

14. See In re Bose Corp., 580 F.3d 1240, 1245 (Fed. Cir. 2009).

15. Kashmir Crown Baking LLC v. Kashmir Foods, Inc., No. 2:12-cv-01780 (KM), 2013 WL 12142539, at *11 (D.N.J. Jan. 18, 2013); see also Edge Games, Inc. v. Elec. Arts, Inc., 745 F. Supp. 2d 1101, 1110 (N.D. Cal. 2010) (denying preliminary injunction where the defendant presented compelling evidence that the plaintiff had submitted doctored specimens to the USPTO to secure its federal trademark registrations); Horizon Healthcare Servs., Inc. v. Allied Nat’l, Inc., No. 03-4098 (JAG), 2006 WL 344277, at *7 (D.N.J. Feb. 14, 2006) (ordering the cancellation of the defendant’s trademark registration because the defendant falsely stated that the specimen was “in use”).

Eric J. Perrott

Eric J. Perrott is a trademark and copyright attorney with Gerben Law Firm, PLLC. He works on trademark prosecution issues and TTAB litigation, while serving as chair of the ABA-IPL’s USPTO Operations Relating to Trademarks Committee.

The material in all ABA publications is copyrighted and may be reprinted by permission only. Request reprint permission here.