Three questions petitioners should ask before filing in 2026

Originally published in The Patent Lawyer Magazine Annual 2026

By R. Parrish Freeman

From the beginning, with their creation in the 2012 America Invents Act (AIA), inter partes review (IPR) and post-grant review (PGR) proceedings skewed against patents and patent owners. “There’s a tension throughout the system,” said then-Chief Judge Randall Rader of the Federal Circuit, speaking at a conference in October 2013, because “you’ve got an agency with 7,000 people giving birth to property rights, and then you’ve got, in the same agency, 300 or so people on the back end... acting as death squads, kind of killing property rights.”

The 300 judges of the Patent Trial and Appeal Board (PTAB or Board) thus found themselves labeled a “patent death squad.” A year later, the Board’s chief judge at the time, James Smith, seemingly embraced the moniker, noting at a public meeting that “[i]f we weren’t, in part, doing some ‘death squadding,’ we would not be doing what the statute calls on us to do.”

And so it went, at least for the next decade. Patent owners could be nearly certain that attempts to enforce their patent rights through litigation would likely land them before the PTAB’s “patent death squad.” The litigation is stayed, or at least derailed, while the challenged patents proceed through IPR or PGR, with no presumption of validity and only a preponderance of the evidence standing between existence and oblivion. Most challenged claims did not survive.

That all started to change last year. Through a spate of policy shifts, new procedures, precedential decisions, and rule-making efforts, the Board has significantly changed the game; an apparent effort to reassure innovators that patent rights are worth pursuing. Patent owners can feel more assured that the litigation won’t be derailed by an IPR or PGR petition, knowing the Director is likely to invoke his discretion to deny the petition for any of a variety of non-substantive “discretionary denial” reasons. Volumes have been written about discretionary denial, and readers’ familiarity with how patent owners can best leverage them is assumed.

The focus of this article is on the now-beleaguered petitioner. As patent infringement defendants, petitioners no longer have that 70% (or so) chance of diverting the invalidity case to the friendly battlefield of the PTAB, where a mere preponderance of the evidence carries the day. In an environment where the Director alone decides petitions and is looking for any reason to deny, petitioners must understand how the landscape has changed and what new questions they need to ask.

Here are three questions to address before assuming an IPR (or PGR) petition is the best path. Some answers may give rise to a decision not to file a petition at all. Others may give the petition a better chance of surviving the Director’s pre-trial gauntlet and proceeding to an IPR or PGR trial before a panel of PTAB judges.

These three questions go to the merit of the petition rather than to any of the well-documented “discretionary denial” factors. The petitioner must confront these three merit-based issues in the petition without waiting for the patent owner to raise them first, because failure to do so could cause the petition to be denied, and a “do-over” petition could be time-barred or suffer discretionary denial as a serial attack.

Arguments against discretionary denial, on the other hand, should not be included in the petition. Under the Directors new bifurcated approach, these are properly raised only in response to a patent owner’s separately-filed request for discretionary denial.

1.          Have you accurately identified all real parties-in-interest and confronted all implicated issues?

In October 2025, the Board announced a new policy – in the form of returning to an old policy – of requiring petitioners to identify the real parties in interest (RPIs) to their petitions before institution. A single omission can now result in denial of your petition, and a second attempt could be time-barred under the one-year limitation of § 315(b).

Clearly identifying all RPIs in the petition is a statutory requirement (§ 312(a)(2)) and has been since the beginning. But RPI disputes can be fact-intensive and involve additional discovery and briefing during the short pre-decision time window required by § 314(b). In 2020, the Board essentially threw up its hands with its decision in SharkNinja Operating LLC v. iRobot Corp., which it designated “precedential” on the subject of RPI disclosure as a petition requirement. The SharkNinja panel held that petitioners need not address, and the Board need not resolve, nettlesome RPI issues unless doing so would result in the petition being time-barred or the petitioner being estopped as a result of its relationship with an unnamed RPI. Last September, the Board removed SharkNinja’s “precedential” designation, stating simply that “disclosure of RPIs serves purposes other than time-bar and estoppel issues.”

In his October 28 memorandum on the subject, Director Squires haled a return to “the Office’s pre-SharkNinja practice of requiring petitioners to identify the real parties in interest (‘RPIs’) to their petitions before institution by designating as precedential Corning Optical Communications RF, LLC v. PPC Broadband Inc., IPR2014-00440, Paper 68 (PTAB Aug. 18, 2015) (except for §II.E.1).” The Corning Optical panel reasoned that “[b]ecause Petitioner here has failed to name all real parties-in-interest, the Petitions have not met the requirements of 35 U.S.C. § 312(a)(2), and, therefore, are incomplete and cannot be considered.” Moreover, the panel noted that “even if Petitioner amended its Petitions, those Petitions would receive a new filing date,” which would result in the petition being time-barred under § 315(b) for filing more than one year after service of a complaint alleging infringement. Your RPI assessment may reveal that your petition is a non-starter due to time-bar or estoppel issues. If it doesn’t, take pains to identify any and all potential RPIs in the petition. The patent owner can defeat the petition just by pointing to an inaccurate or incomplete RPI disclosure. And, unlike in years past, curative efforts will result in a new and possibly untimely filing date.

2.          Is the petition claim construction any different than the construction proposed in parallel litigation?

The Board expects alignment – or a reasoned explanation – when claim construction positions differ from district court litigation. In November 2025, the Director issued a precedential decision in Revvo Techs., Inc. v. Cerebrum Sensor Techs., Inc., requiring petitioners explain why the differences are warranted, and rejecting a panel decision that would have limited the requirement to instances involving § 112(f) means-plus-function language. Rule 42.104(b)(3), 37 C.F.R., requires an IPR petition to identify “[h]ow the challenged claim is to be construed,” so an unexplained discrepancy with a concurrently-urged district court construction amounts to a failure to meet this requirement. Just two days after Revvo, the Director issued a decision in Tesla, Inc. v. Intellectual Ventures II LLC, designated “informative,” to clarify that the requirement applies even where the petitioner professes agreement with the patent owner’s position for purposes of deciding the petition, despite simultaneously arguing for a particular construction in the district court, or that the language is indefinite and cannot be construed. The Director denied the petition because “[a]llowing a petitioner to advance a claim construction before the Board when that petitioner has made inconsistent indefiniteness arguments in district court fails to further, but instead detracts from, the Office’s goal of ‘providing greater predictability and certainty in the patent system.’”

More so now than ever, your petition should note any claim construction positions taken in a parallel district court action and explain anything that could appear as a discrepancy to the Board. Failure to do so risks outright denial of the petition.

3.          For IPR petitions, have you identified all the claim elements in either a patent or a printed publication, with no reliance on AAPA or POSITA common knowledge to supply a missing claim element?

Your IPR petition must identify all elements of the challenged claims in prior art patents or printed publications, without reliance on the skilled artisan’s “general knowledge” or on “applicant admitted prior art” (AAPA) derived from statements in the challenged patent about the content of the prior art. This is a recent change and one that is somewhat more restrictive than what the Federal Circuit has said. Failure to comply will result in the petition being denied.

In 2022, the Federal Circuit held in Qualcomm Inc. v. Apple Inc. that AAPA cannot form the “basis” of a petition’s stated ground for unpatentability, and clarified in a 2025 decision in the same case that AAPA forms such a basis even if identified “in combination” with a patent or printed publication. Later, in July 2025, the Acting Director issued a memorandum announcing that the USPTO “will enforce and no longer waive the requirement of 37 CF.R. § 42.104(b)(4) (Rule 104(b)(4)) that a petition for inter partes review (IPR) ‘must specify where each element of the claim is found in the prior art patents or printed publications relied upon.’” “As a practical matter,” the memorandum continues, “enforcement of Rule 104(b)(4) means that applicant admitted prior art (AAPA), expert testimony, common sense, and other evidence that is not ‘prior art consisting of patents or printed publications’ (collectively, ‘general knowledge’) may not be used to supply a missing claim limitation.” To avoid doubt, the memorandum states unequivocally that, “[p]er 35 U.S.C. § 312(a)(4), the Board shall deny an IPR petition that fails to comply with Rule 104(b)(4).”

The Board cannot grant an IPR petition that is not in full compliance with 35 U.S.C. § 312(a), “Requirements of Petition.” Subparagraph (4) requires that “the petition provides such other information as the Director may require by regulation,” and one such regulation is Rule 42.104(b)(4), which states that “[t]he petition must specify where each element of the claim is found in the prior art patents or printed publications relied upon.” On this logic, and in view of the “Director’s broad discretion to prescribe regulations to ensure the ‘integrity of the patent system, the efficient administration of the Office, and the ability of the Office to timely complete proceedings,’” the July 2025 memorandum goes slightly farther than Federal Circuit authority might require. The Acting Director acknowledged this point, but maintained that “enforcing Rule 104(b)(4) is the best course of action to provide certainty to the parties, the Board, and the public, and to allow for the efficient administration of the Office.”

While the Federal Circuit’s holdings remain opaque as to what amounts to forming a “basis” of a ground for unpatentability, the July 2025 policy announcement clarifies that all elements of the challenged claims must be identified as either expressly disclosed or implied by teachings found in prior art patents or printed publications. The petition will be denied if this requirement is overlooked, which could ripple into time-bar issues for a second attempt, or open up discretionary denial factors not present at the time of the first petition.