by Dennis Crouch
Samsung and the U.S. government have filed briefs urging the Supreme Court to deny certiorari in Lynk Labs, Inc. v. Samsung Electronics Co., No. 25-308. At issue is whether a patent application filed before a challenged patent’s priority date, but not published until afterwards, can serve as prior art in inter partes review. The Federal Circuit held that it can, applying the filing-date rule of pre-AIA 35 U.S.C. § 102(e)(1). As I write below, I was not overly impressed with they briefing.
The case centers on the “Martin” reference, a patent application covering LED technology that was filed on April 16, 2003 and published on October 21, 2004. Martin was later abandoned and never became a patent. Lynk Labs’ ‘400 patent claims a priority date of February 25, 2004, placing it squarely in the gap between Martin’s filing and publication dates. Samsung successfully used Martin to challenge claims of the ‘400 patent as obvious in IPR.
As part of its streamlining effort, Congress limited what prior art can be used in inter partes review proceedings – only “prior art consisting of patents or printed publications.” 35 U.S.C. § 311(b). Although Martin was not publicly accessible when Lynk Labs filed its application, Martin was later published and also qualifies as prior art under pre-AIA 102(e)(1) (or its parallel post-AIA statute 102(a)(2)). In reading the statute, the court concluded that Martin can be used in an IPR because it satisfies each of the two requirements (prior art + printed publication) even though its prior art date is well before its publication date. Lynk’s cert petition frames this as a question about whether “printed publications” in § 311(b) carry the traditional public-accessibility timing requirement, or whether a published patent application can be backdated as prior art to its filing date in IPR.
I have previously written about this case at several stages. Dennis Crouch, Secret Springing Prior Art and Inter Partes Review, Patently-O (Oct. 2024); Dennis Crouch, Publications Before Publishing and the Federal Circuit’s Temporal Analysis, Patently-O (Jan. 2025); Dennis Crouch, Prior Art Document vs. Prior Art Process, Patently-O (July 2025); Dennis Crouch, Supreme Court Asked to Resolve “Secret Springing Prior Art” in Inter Partes Review Proceedings, Patently-O (Sept. 2025); Dennis Crouch, Lynk Labs: How the Least-Vetted Documents Destroy Issued Patents, Patently-O (December 2025).
Before getting into the merits of the responses, I first want to note how both briefs appear shockingly disingenuous in their discussion of the limited impact of the decision.
Samsung devotes pages arguing the case is unworthy of review because it arises under a “superseded” pre-AIA statutory scheme, while simultaneously conceding that the current post-AIA statute contains an essentially identical provision. The old 102(e) is now codified as 102(a)(2), but is virtually identical for the purposes of this case.
Pre-AIA 102(e)(1): A person shall be entitled to a patent unless the invention was described in an application for patent, published under section 122(b), by another filed in the United States before the invention by the applicant for patent
Post-AIA 102(a)(2): A person shall be entitled to a patent unless the claimed invention was described … in an application for patent published … under section 122(b) [naming] another inventor and was effectively filed before the effective filing date of the claimed invention.
Samsung suggests that the court should wait for a post-AIA case. But the analysis effectively concedes that the statutory question is not limited to a dwindling set of pre-AIA patents. The SG’s brief makes the concession even more explicit. A footnote reads: “for purposes of the question presented here, however, there are no material differences between the pre- and post-AIA versions of Section 102.”
The Solicitor General’s brief (signed by USPTO officials) makes its own misguided statements that the impact of the case is “narrow” since it only involves prior art filed during an 18-month lookback period. The U.S. government argues:
[T]he question presented is of limited practical importance, since it will arise only in an IPR where the challenged patent’s priority date falls within the window (ordinarily 18 months) between the filing and publication of a patent application that the IPR petitioner invokes as prior art.
US Brief at 7. But patent attorneys know that this 18-month window is often incredibly important — particularly in technology sectors experiencing rapid growth, which are also precisely the areas most likely to generate litigation and IPR challenges. The government’s characterization of the issue as “narrow” ignores the practical reality of patent prosecution and validity disputes. A quick survey of PTAB decisions issued this week illustrates the point: of eight decisions I looked at involving § 103 obviousness, half relied upon § 102(a)(2) application prior art. The USPTO is increasingly leveraging these unpublished-at-the-time references to invalidate patents, and the 18-month gap between filing and publication represents prime territory for prior art in fast-moving fields. Dismissing this as marginal ignores how modern patent law actually unfolds.
"Lion is a Mammal:" The Federal Circuit used an analogy: just as a lion is a specific type of mammal, a published patent application is a specific type of printed publication. Section 311(b)'s reference to "printed publications" thus encompasses published patent applications, along with the timing rule Congress attached to them in § 102(e)(1). Both respondents rely on this framing.
Both briefs in opposition treat the primary question as whether a published patent application is a “printed publication” for purposes of § 311(b), and that answer is plainly “yes.” But this framing obscures the separate question about what makes the application prior art in the first place.
Under § 102(e)(1), the prior art is the application, not the publication. The statute’s text makes this clear: a patent is barred if the invention was “described in an application for patent [later] published under section 122(b).” The subject of the sentence is “an application for patent.” The key trigger is the filing date and publication is a condition that must be satisfied for the application to have prior art effect. But, it is the act of filing, not the act of publication, establishes the prior art date. The application is the prior art; publication is what activates it.
This distinction matters for the § 311(b) question. If a published patent application derives its prior art status from being an application filed before the invention, then it functions as prior art in a way that is categorically different from how a book, journal article, or any other printed publication functions.
Indeed, re the false briefing re the alleged non-importance of the prior art [and prior invention] of a published application being prior art as of its filing date.
However, on the merits, there is no statutory, prior-practice, or logical prior art difference whether a published application also gets published a second time later as a patent or not. Nor would it help international patent law uniformity. With the opposition to cert of the Solicitor that should end that novel argument.
You don’t explain why you include the statement “Nor would it help international patent law uniformity.” My best guess is that you are indicating that your disagree with an observation Prof. Crouch made in this area as far back as 2010. In any event I think you are not correct. Reversing in Lynk Labs would help international patent law uniformity, by eliminating one of U.S. law’s longstanding roadblocks to such harmonization.
As Professor Crouch noted years ago (‘Federal Circuit Extends the Scope of 102(e) “Secret Prior Art”’, Patently-O, July 7, 2010, link to patentlyo.com, allowing the use of secret prior art to show obviousness, as Lynk Labs does, is the opposite of Europe’s approach. Europe does allow a limited use of so-called secret prior art, but only with respect to novelty, not inventive step—thus, corresponding to U.S.-style anticipation, but not obviousness. (The European Boards of Appeal’s recent G 1/23 decision arguably took a tiny step in the U.S.’s direction, but only with respect to the state of the art consisting of analyzable chemical compositions of products in prior public use, and it simply clarified that there needn’t be shown to have been a specific prior reason to conduct such an analysis prior to the later invention.)
The reason Europe gives for its no-secret-prior-art-for-inventive-step rule is straightforward: such prior art for novelty is necessary to reduce double-patenting, but allowing it for inventive step (i.e., obviousness) is an unreasonable extension of the prior art when later inventors can’t possibly have known about it at the time of their later invention. (See, e.g., Michael Caine, The Problem With Secret Prior Art (Davies Collison Cave, Nov. 17, 2017), link to dcc.com; see also Prof. John R. Thomas, The Scope of the Prior Art, 93 GWU Law Rev. 54 (2025).)
In harmonization talks, the U.S.’s oddball position that secret prior art can show obviousness has reportedly long been one of the biggest roadblocks, because Europe quite reasonably does not want to reduce the scope of patentability in an unreasonable manner for the sake of harmonization. (See, e.g., Caine, supra.)
To be sure, reversing Lynk Labs won’t create complete harmonization: it won’t change the fact that U.S. patent applications that actually are granted as patents will still be available in the U.S. as prior art to show obviousness, which will remain true as long as the U.S. Supreme Court’s judge-made rule in Hazeltine remains the law. But it would at least eliminate the expansion of such obviousness prior art to secret abandoned and unexamined applications.
Yes. This would go part of the way toward international harmonization.