The law for design patent obviousness may change in the upcoming months when a decision is issued from the Federal Circuit Court of Appeals from the February 2024 en banc rehearing[1] of LKQ Corp. v. GM Global Technology Operations LLC.[2] Although the law has not yet changed, the application of the current law has recently changed at the United States Patent & Trademark Office (USPTO). With the use of AI-assisted searching, design patent examiners appear to be issuing more obviousness rejections under the current law. While we await any change to the law, practitioners should be prepared to address AI-assisted design patent rejections, and should develop strategies for maintaining a design patent portfolio while the law and procedure evolve.
The current test for US design patent obviousness is known as the Rosen-Durling test. Under this test, “one must find a single reference, ‘a something in existence, the design characteristics of which are basically the same as the claimed design'”[3], often referred to as a ‘Rosen reference’. Secondary references “may only be used to modify the primary reference if they are ‘so related to the primary reference that the appearance of certain ornamental features in one would suggest the application of those features to the other.'”[4]
LKQ Corp. challenged GM’s US Design Patent D797,625 S for a Vehicle Front Fender in an Inter Partes Review before the Patent Trial and Appeal Board (PTAB) of the USPTO.[5] The patent was challenged over BYD Co. Ltd., Vehicle USD773,340 S, alone and in view of a Hyundai Tucson brochure. The PTAB concluded that LKQ Corp. failed to identify a Rosen reference with basically the same visual impression of the patented design.
LKQ appealed the PTAB decision to the Federal Circuit Court of Appeals to determine whether the Rosen-Durling test was implicitly overruled by the Supreme Court decision in KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398 (2007), which permitted flexibility in considering obviousness in the context of a utility patent. A three-judge panel decided that KSR did not overrule the Rosen-Durling test.[6] LKQ requested an en banc rehearing which was conducted before a ten-judge panel in February 2024.[7]
LKQ argued that the Rosen-Durling test for design patent obviousness is more rigid than the teaching-suggestion-motivation test for obviousness, which was loosened by the KSR decision. LKQ also argued that the analogous arts test is a more appropriate test for design patent obviousness. The USPTO agreed that the Rosen-Durling test could be loosened. GM argued that the Rosen-Durling test has been effective at invalidating obvious design patents for over forty years and is not overly rigid.
The hearing addressed various issues without a clear indication of which direction the court may decide. The judges discussed whether a prior art reference in design patent obviousness must be the same article of manufacture as the claimed design, which is a requirement in design patent anticipation[8] and infringement.[9] The court expressed the difficulty in applying the same obviousness case law to utility patents and design patents. Questions were raised as to what problem is solved by a design patent in the context of obviousness for analogous arts. One judge asked whether a simple substitution would jeopardize most design patents. Another judge pointed out that the parties did not propose a test to replace the Rosen-Durling test. The court also expressed a concern as to whether this issue was industry specific. During the hearing, the USPTO indicated that four percent of design patent applications are rejected as obvious.
Over the upcoming months, we will await a decision from the en banc rehearing. It is unpredictable whether the court will substitute the Rosen-Durling test because no substitute test was proposed by the parties. The Rosen-Durling test may be loosened to be applied more flexibly. Or the court may reaffirm the decision from last year without modifying the design patent obviousness test.
In the meantime, the USPTO recently employed artificial-intelligence-assisted prior art searching.[10] Design patent file histories are only accessible to the public upon issuance, and therefore we have limited data at this time to assess the outcome and statistics of these searches and the resulting rejections. From the perspective of my docket, which is a limited sample set, the AI-assisted prior art searches represent an increase in quality of the prior art searches. The Examiners are applying prior art references from a wider variety of sources, including an increase in rejections over non-patent literature and various international industrial design registrations or patents. In recent months, the majority of first office actions that I received in design patent applications include an obviousness rejection applying the Rosen-Durling test. In other words, the allegedly ‘rigid’ test for applying non-obviousness is being applied in a majority of recent office actions, resulting in rejections at a much greater rate than the USPTO-indicated rate of four percent.
Similar to utility obviousness rejections, one should be cautious with obviousness rebuttals to avoid unnecessary file history estoppel, and to avoid surrendering equivalents. A practitioner may prefer to argue differences in the overall visual impression before arguing specific novel features of the claimed design to balance these risks. The practitioner may reserve arguments to specific novel features for a second or subsequent response.
Non-patent literature often discloses limited views of the prior art reference. Arguing the overall visual impression can be effective in rebutting rejections over non-patent literature since the non-patent literature is often missing multiple views of the claimed design.
An obviousness rejection often depicts the claimed images adjacent to the prior art references for comparison. The reproduction of non-patent literature and various foreign references are often pixelated and blurry, especially if the source image was in color and converted to gray scale for the rejection. Researching the source images may obtain clearer images of the prior art reference. By comparing clearer images of the prior art references to the claimed images in a response, a practitioner may be more persuasive in arguing and contrasting differences in the overall visual impression that were otherwise difficult to observe in the gray scale reproductions.
In view of the AI-assisted searching and the possible upcoming changes to design law, one should manage expectations so that applicants are prepared to face an increase in obviousness rejections. In the context of international portfolios, many foreign equivalents may have already granted by the time the US obviousness rejection is issued. Patience and persistence may be required to obtain similar protection in the US.
Various strategies may be employed in the preparation of design patent applications to bolster the applications in advance of the developing examination procedures. It may be helpful to conduct novelty searches with AI-assisted image based searching tools to locate relevant references before preparing the patent application. If a design is within a crowded field, it may be advantageous to focus on key design features in the design patent application.
A difficulty in design patent preparation is reducing a three-dimensional design to two-dimensional images. Distinguishing features, such as contours, may be highlighted with sectional views to help distinguish over the prior art.
Conversely, sectional views may risk adding narrowing limitations to a design patent application. A practitioner may consider filing the design patent application with multiple embodiments: with, and without a sectional view(s). If a restriction requirement is received, these office actions often include a list of references cited by the Examiner. The applicant may study the cited art to determine which embodiment should be elected.
Design practitioners may also consider whether to file the design with line drawings or rendered computer-aided design (CAD) figures. Rendered CAD views can sometimes illustrate contours and shapes more effectively than line drawings.
These strategies have been developed to address the current state of the law of design patent obviousness, to address recent changes to the application of the laws by employing AI tools, and to prepare for future changes to the law and procedures. As the law and procedures continue to evolve, further strategies may be developed.
Please note that the views, opinions, experiences, suggestions, and recommendations provided herein solely represent those of the author, and do not represent views or opinions of the author’s employer or any entities they represent, or the publisher.
[1] LKQ Corp. v. GM Global Tech. Operations, No. 2021-2348 (Fed. Cir. 2024).
[2] LKQ Corp. v. GM Global Tech. Operations, 71 F.4th 1383 (Fed. Cir. 2023).
[3] Durling, v. Spectrum Furniture Co., Inc., 101 F.3d 100, 103, (Fed. Cir. 1996) quoting In re Rosen, 673 F.2d 388 (CCPA 1982).
[4] Durling, 101 F.3d at 103, quoting In re Borden, 90 F.3d 1570, 1575 (Fed. Cir. 1996).
[5] LKQ Corp. v. GM Global Tech. Operations, IPR2020-00534 (PTAB 2021).
[6] LKQ v. GM, 71 F. 4th 1383.
[7] LKQ v. GM, No. 2021-2348.
[8] In re SurgiSil, LLP, 14 F. 4th 1380 (Fed. Cir. 2021).
[9] Curver Luxembourg v. Home Expressions, 938 F. 3d 1334 (Fed. Cir. 2019).
[10] Manual of Patent Examining Procedure § 901.06(a)(IV)(B), (Rev. 07.2022, February 2023).
