Returning IPR Merits Determinations to the PTAB
Executive Summary: The announced return of merits institution determinations in inter partes review to the Patent Trial and Appeal Board offers an opportunity to align decisional responsibility with institutional expertise. Because the Director-centered process already incorporates administrative patent judge consultation, the case for reform rests on giving specialist panels responsibility for collective deliberation, resolving technical disagreements, and explaining their decisions. The Director’s comparative advantage lies in establishing and supervising lawful Office-wide discretionary policy, with room to delegate routine applications. Many considerations underlying discretionary denial—including trial schedules, litigation investment, party identity, and settled expectations—can be evaluated without resolving how the patented technology operates. Patentability assessments, by contrast, require integrated legal and technical analysis suited to APJs’ statutory qualifications and professional experience. This distinction remains functional: mandatory requirements such as the statutory time bar may be nontechnical, while discretionary review may depend on technical assessments. A transparent system of reasoned APJ recommendations and explained Director departures could achieve many of the same benefits, although direct panel decisions offer clearer responsibility and fewer routine handoffs. Implementation should also resolve whether the institution panel will conduct the ensuing trial, weighing efficiency and technical familiarity against plausible risks of anchoring and confirmation bias. The reform’s success should ultimately depend on reasoned decisions, accountable supervision, appropriate technical assignments, and genuine reconsideration as the evidentiary record develops.
I The case for returning technical decisions to specialist judges
Returning merits institution determinations in inter partes review to panels of administrative patent judges should establish a clearer allocation of decisional responsibility within the United States Patent and Trademark Office. The Director-centered process already incorporates APJ expertise. The proposed reform should make specialist panels responsible for reaching and explaining the ordinary merits determination, while preserving accountable agency supervision. The Director’s comparative advantage lies in establishing and supervising lawful Office-wide discretionary policy. APJs’ comparative advantage lies in resolving case-specific patentability disputes through integrated legal and technical analysis. These responsibilities permit practical delegation and review; neither requires every institution question to be decided personally by the Director.
The announced change provides an opportunity to put that division into practice. In his September 29, 2026 written testimony, Director John A. Squires stated his intention to return merits institution determinations to the Patent Trial and Appeal Board in October, approximately one year after centralizing them. His testimony linked the intended return to training and institutional development through PTAB STRONG. Oversight of the United States Patent and Trademark Office: Hearing Before the Subcomm. on Intell. Prop. of the S. Comm. on the Judiciary, 119th Cong. 7 (2026) (statement of John A. Squires, Under Sec’y of Com. for Intell. Prop. & Dir., U.S. Pat. & Trademark Off.) [hereinafter Squires Testimony].
That announcement follows two different allocations of institution responsibility. In March 2025, the Office separated discretionary review by the Director from merits and other nondiscretionary issues decided by a three-member PTAB panel. In October 2025, the Director assumed responsibility for both stages, with consultation from at least three APJs. That process contemplated routine summary notices and allowed referral to Board members when detailed treatment was appropriate. Memorandum from Coke Morgan Stewart, Acting Under Sec’y of Com. for Intell. Prop. & Acting Dir., U.S. Pat. & Trademark Off., to All PTAB Judges, Interim Processes for PTAB Workload Management 1–3 (Mar. 26, 2025) [hereinafter March 2025 Memorandum]; Memorandum from John A. Squires, Under Sec’y of Com. for Intell. Prop. & Dir., U.S. Pat. & Trademark Off., to All PTAB Judges, Director Institution of AIA Trial Proceedings 1–2 (Oct. 17, 2025) [hereinafter October 2025 Memorandum].
The institutional case for returning merits decisions to APJs depends on what responsibility adds to consultation: collective panel deliberation, resolution of disagreements among specialists, a reasoned explanation attributable to the judges reaching the decision, and sustained attention distributed across the Board’s panels. Those are proposed mechanisms for improving adjudication, not established consequences of changing the formal issuer. A Director-centered system that ordinarily adopts a published, reasoned APJ recommendation and explains departures is a serious alternative. Comparing those arrangements makes the case for direct panel decisionmaking more precise. The September announcement describes an intended transition; implementing procedures will determine whether it realizes these advantages and how institution panels relate to later trial panels.
II Specialist consultation and decisional responsibility
Because the centralized process already draws on APJs, the case for reform must explain what direct panel responsibility would change. The October 2025 memorandum required consultation with at least three judges, contemplated summary institution notices, and permitted referral of issues warranting detailed treatment to Board members. October 2025 Memorandum, supra, at 1–2. Those provisions established access to specialist judgment. They did not expressly require a jointly reasoned APJ recommendation to be published in every ordinary case or identify how disagreements among consulting judges would be resolved. The proposed advantage of direct panel adjudication concerns the organization and explanation of that specialist work.
Collective deliberation is the first mechanism. A panel charged with deciding a petition should require its members to test competing interpretations of the claims and references against the record and resolve material disagreements about the challenge. If one judge sees a disclosed function as satisfying a limitation and another reads the reference differently, the panel must confront that difference in reaching its judgment. Direct responsibility makes resolution of the disagreement part of the specialists’ adjudicative task. Consultation can also involve rigorous collective deliberation, and nothing in the memorandum establishes that consulting APJs failed to deliberate. The recommendation is to make that deliberation an express feature of the ordinary decisional process.
A reasoned explanation attributable to the deciding panel is the second mechanism. The judges who reach the merits determination should explain the technical propositions and legal reasoning on which it depends. That obligation can expose an unsupported inference, require an answer to significant contrary evidence, and make disagreements visible when a judge writes separately. It also allows the parties and reviewing officials to connect the outcome to the panel’s treatment of the record. These advantages arise from a reasoned and identifiable decision, not from replacing one official’s signature with another. A summary notice issued in a panel’s name would provide little of the claimed benefit.
Allocation of sustained attention is the third mechanism. Appropriately assigned panels can take responsibility for developing and completing case-specific merits analyses across the Board’s docket, while the Director concentrates supervisory attention on consequential legal or institutional questions. This arrangement can avoid routinely passing a completed technical assessment to another decisionmaker for an additional approval step. Its value depends on staffing, workload, and the actual review process; delegation does not create capacity by itself. Sustained attention at institution also does not require retaining the same individuals for the later trial, a separate design choice addressed below.
“Generalist” describes the breadth of the Director’s institutional responsibilities. It does not imply that a particular Director lacks patent experience or scientific understanding, or that consulting APJs cease to be specialists. The relevant comparison concerns the allocation of decisional work within an agency that possesses expertise under either arrangement. Direct panel responsibility is preferable to the extent that it gives the specialists a clear obligation to reach and explain the technical judgment while reserving the Director’s broader oversight for questions warranting supervisory attention.
A Director-centered system built around reasoned APJ recommendations is a substantial alternative. Three APJs could deliberate collectively, prepare a written recommendation identifying their reasoning and disagreements, and have that recommendation published when the Director adopts it. The Director could publicly explain any departure and return unresolved technical questions to the specialists. Such a process could preserve technical rigor and provide an additional opportunity to identify inconsistent legal treatment across panels. It would address many of the concerns associated with routine summary notices, even though the Director remained the formal decisionmaker.
The preference for direct panel decisions is therefore narrower than a claim that recommendations are inherently inferior. Where the panel has already produced a complete, reasoned determination, requiring a further decision in every ordinary case may add a handoff without a corresponding improvement in analysis. It can also leave responsibility for particular reasons divided between those who developed them and the official who adopted the outcome. Direct panel adjudication gives the specialists primary responsibility for the merits judgment and treats supervisory intervention as a defined review function. The Director remains able to address legal inconsistency and other matters within lawful supervisory authority without routinely becoming a second issuer of every technical determination.
That preference must remain conditional. If a recommendation system achieves the same collective deliberation, public attribution, timely resolution, and reasoned supervision without meaningful duplication, its practical performance may approach that of direct panel adjudication. Conversely, poorly staffed panels issuing unexplained decisions would not realize the advantages proposed here. The mechanisms described are institutional reasons to favor a particular default allocation, not empirical proof that changing formal authority necessarily improves accuracy. Evaluation should examine the quality of reasons, handling of disagreement, timeliness, and the operation of review.
Continued Director supervision is compatible with primary APJ responsibility. United States v. Arthrex addressed the Director’s ability to review final APJ decisions and provided a remedy allowing such review. It did not require the Director personally to adjudicate every institution petition. United States v. Arthrex, Inc., 594 U.S. 1, 25–27 (2021) (opinion of Roberts, C.J.). Although that holding concerns final decisions rather than the precise institution procedure proposed here, it illustrates that specialist adjudication can coexist with accountable agency supervision. The task is to specify both responsibilities and how they interact.
III The scope of the discretionary and merits distinction
The institution process involves decisions of different kinds. Discretionary review principally asks whether the Office should devote its adjudicative resources to a particular challenge in light of procedural history and applicable policy. Merits review asks whether the petitioner’s legal and technical showing meets the statutory threshold. Section 314(a) requires a reasonable likelihood that the petitioner would prevail as to at least one challenged claim before institution may be authorized. Satisfaction of that threshold does not itself require institution. 35 U.S.C. § 314(a); SAS Inst. Inc. v. Iancu, 584 U.S. 357, 365–66 (2018).
Many discretionary considerations are nontechnical in their essential character. The significance of prior proceedings, settled expectations, duplication, and the Office’s workload generally turns on institutional judgment rather than scientific analysis. The March 2025 memorandum expressly identified these types of considerations and linked its procedures to the Board’s ability to meet competing workload obligations. March 2025 Memorandum, supra, at 2–3. Whether a proceeding would duplicate work already performed can often be evaluated without deciding whether a particular circuit, chemical composition, or mechanical arrangement satisfies a claim limitation.
Technical complexity and the legal character of an institution requirement are separate dimensions. Nontechnical does not mean discretionary, and technical does not mean merits-only. Section 315(b), for example, bars institution when the petition is filed more than one year after service of an infringement complaint on the petitioner, a real party in interest, or its privy, subject to the statute’s exception for a request for joinder. 35 U.S.C. § 315(b)–(c). Applying that limitation may turn on service dates and party relationships without requiring an understanding of the invention; it is nevertheless a statutory restriction, not an invitation to balance administrative convenience. Conversely, the March 2025 memorandum permits consideration of challenge strength and expert reliance within discretionary review, potentially requiring substantial technical judgment. March 2025 Memorandum, supra, at 2. The proposal principally allocates discretionary policy judgments and technical patentability assessments. Other statutory institution requirements need an expressly designated decisionmaker and procedure; their nontechnical character does not assign them automatically to the Director.
The practical importance of this distinction appears in the Office’s June 2026 statistics. For fiscal year 2026 through June, the report shows a 38 percent overall institution rate by petition, a 60 percent denial rate within the discretionary-determination category, and a 75 percent institution rate within the merits-and-other-determination category. These figures combine IPR and post-grant review, and the categories are not a single matched cohort. They nevertheless illustrate the substantial role of discretionary screening. U.S. Pat. & Trademark Off., PTAB Trial Statistics June 2026 6–7 (2026). Returning merits decisions to APJs can improve the quality of that stage without necessarily producing a comparable change in overall access to trial.
IV Director responsibility for discretionary policy
The affirmative case for Director responsibility begins with the scope of the office, not with the absence of a need for technical training. APJs can evaluate scheduling orders, litigation history, commercial records, and reliance. The Director’s distinct responsibility is to establish lawful institutional policy across proceedings, reconcile competing demands on the Office, and supervise consistent application. Congress directed the Director, when prescribing IPR regulations, to consider the economy, the integrity of the patent system, efficient administration, and the Office’s ability to complete proceedings on time. 35 U.S.C. § 316(a)–(b). That provision does not resolve every dispute about the lawful reach of discretionary denial, but it identifies considerations that require an Office-wide perspective.
Individual panels see the circumstances of the cases assigned to them. Agency leadership is positioned to consider those circumstances alongside aggregate caseloads, competing statutory duties, and the consequences of applying a policy throughout the system. A uniform approach to parallel litigation or settled expectations also requires choices about the relative weight of competing interests. The Director should articulate those choices within the governing law, explain policy changes, and remain answerable for how the Office administers them. Central responsibility for policy can make those choices more visible and consistent than leaving each panel to formulate its own approach.
Establishing policy and personally applying it in every proceeding are different functions. A Director may prescribe a uniform approach to parallel litigation while delegating routine findings about trial dates, party identity, or procedural history to APJs. Delegation can preserve case-specific attention and avoid a central approval queue, provided the policy is clear and meaningful supervisory review remains available. The Director can retain novel policy questions, consequential departures, and decisions about agency resources that individual panels cannot resolve in isolation. This is a proposal for accountable policy leadership with practical case-level delegation, rather than a claim that APJs are less capable of deciding nontechnical facts.
The discretionary criteria below identify why this division is workable. Many can be applied from scheduling orders, litigation filings, patent issuance dates, correspondence, and business records. Their predominantly nontechnical character supports separating them from the substantive patentability analysis. The affirmative reason for Director leadership remains uniform policy and institutional accountability; the factual application may be performed by the Director or by APJs acting under that policy.
A The Fintiv factors principally address the management of parallel proceedings
The six factors identified in Apple Inc. v. Fintiv, Inc. address stays, relative timing, investment in parallel litigation, overlapping issues, party identity, and other circumstances, including the merits. They operate as a holistic assessment of whether a second proceeding would advance efficiency and the integrity of the adjudicative system. Apple Inc. v. Fintiv, Inc., No. IPR2020-00019, Paper 11, at 6 (P.T.A.B. Mar. 20, 2020) (precedential). Examining each factor shows that most of the framework concerns the relationship between proceedings rather than the operation of the patented technology.
The first factor concerns whether the court has stayed the parallel litigation or whether evidence indicates that it may do so following institution. Id. at 6–9. The relevant materials are the stay order, any pending motion, the court’s stated reasons, and the procedural circumstances. A decisionmaker can determine that a stay would reduce simultaneous litigation without understanding a semiconductor’s architecture or a pharmaceutical compound’s properties. Predicting a future stay may involve uncertainty, but uncertainty about a court’s procedural course is not a technical patentability question.
The second factor compares the court’s anticipated trial date with the Board’s projected deadline for a final written decision. Id. at 9. This inquiry requires evaluating calendars and the reliability of scheduling evidence. Whether the district court is likely to reach trial before the PTAB completes its review can be assessed without construing the invention’s technical operation. Scheduling orders, continuances, and evidence of the court’s actual pace matter because they bear on duplication and relative timing. Technical training does not supply the central competence needed to weigh that evidence.
The third factor concerns the work already invested in the parallel proceeding and the petitioner’s diligence in seeking review. Id. at 9–12. The adjudicator can examine whether the court has issued a claim-construction order, how far discovery has progressed, and whether substantial work would be duplicated. Recognizing that a court has construed disputed claims is different from deciding whether its construction is correct. Likewise, assessing the interval between learning which claims are asserted and filing an IPR petition ordinarily requires a chronology and an explanation for delay. These are familiar litigation-management judgments that do not depend on resolving the underlying technical dispute.
The fourth factor examines overlap between the issues in the petition and the parallel proceeding. Id. at 12–13. Much of that comparison can be made from the parties’ pleadings and invalidity contentions: which claims are challenged, which references and combinations are asserted, and which arguments would remain in court after a stipulation. Identifying an identical challenge in two filings does not require deciding whether that challenge succeeds. The important qualification is that substantive equivalence can be disputed. If the question is whether differently worded claims or apparently different references present materially the same technical issue, specialist assistance may be necessary. The administrative consequence of duplication remains distinct from the technical comparison needed to establish duplication in such a case.
The fifth factor asks whether the petitioner is the defendant in the parallel proceeding. Id. at 13–14. Party identity and relevant relationships can ordinarily be established through pleadings, corporate records, and litigation history. Even when those relationships are contested, the dispute concerns who is litigating and the consequences of parallel proceedings. It does not ordinarily require an understanding of the invention’s circuitry, chemistry, software, or mechanics.
The sixth factor encompasses other circumstances, expressly including the merits. Id. at 14–16. Its merits component is the principal limit on describing the entire framework as nontechnical. Deciding that a reference fails to disclose a limitation, or that a proposed combination lacks a persuasive rationale, may require substantial technical understanding even at a preliminary stage. This component should draw on APJ expertise. Its inclusion does not transform the other factors into technical questions: a merits assessment can inform the overall discretionary judgment while the discretionary decisionmaker applies the Director’s lawful policy to the procedural and institutional considerations.
The intervening policy changes reinforce the need to separate the criteria from their particular weighting. The Office rescinded its June 2022 guidance in February 2025. The March 24, 2025 memorandum restored reliance on the earlier precedential framework, included parallel International Trade Commission proceedings, treated timely Sotera stipulations as highly relevant but not independently dispositive, permitted trial-timing evidence including median statistics, and explained that compelling merits alone do not control. Memorandum from Scott R. Boalick, Chief Admin. Pat. J., to Members of the Pat. Trial & Appeal Bd., Guidance on USPTO’s Recission of “Interim Procedure for Discretionary Denials in AIA Post-Grant Proceedings with Parallel District Court Litigation” 1–3 (Mar. 24, 2025). These choices govern how the Office weighs procedural efficiency against competing considerations. They do not make calculating trial dates or identifying litigants a scientific inquiry.
The resulting institutional distinction is concrete. A generalist decisionmaker can assess whether two tribunals are pursuing the same dispute on overlapping schedules, whether the parties have already invested substantially in one forum, and whether a stay or stipulation will reduce duplication. A specialist panel is better equipped to determine whether the prior art actually teaches the claimed invention. The first set of tasks can often be completed without resolving the second.
B Settled expectations concern chronology and reliance
Settled expectations provide a particularly clear example of a discretionary consideration whose central inquiry is nontechnical. The March 2025 memorandum identified the parties’ settled expectations, including the length of time the claims have been in force, as relevant to discretion. March 2025 Memorandum, supra, at 2. The Director’s subsequent precedential decision in Magnolia expressly recognized that either party’s expectations may matter and identified Dabico and Home Depot as illustrations. Magnolia Med. Techs., Inc. v. Kurin, Inc., No. IPR2026-00097, Paper 17, at 8–9 (P.T.A.B. May 14, 2026) (precedential Director decision). The question concerns the reasons for reopening patentability through an administrative proceeding in light of the parties’ history and expectations.
In Dabico, the Director relied on the patent’s nearly eight years in force and the petitioner’s failure to offer persuasive reasons why review would be an appropriate use of Office resources. The decision rejected a fixed age cutoff and explained that actual notice of the patent or possible infringement was unnecessary to create settled expectations. Dabico Airport Sols. Inc. v. AXA Power ApS, No. IPR2025-00408, Paper 21, at 2–3 (P.T.A.B. June 18, 2025) (informative Director decision). That reasoning can be evaluated by identifying the issuance date, reconstructing the relevant chronology, and assessing the petitioner’s justification for review. None of those steps inherently requires understanding the mechanism disclosed in the patent. Actual knowledge may be relevant on a particular record, but it should not be presented as a universal prerequisite.
Home Depot demonstrates why the analysis must remain attentive to both sides. Although the patent had been in force for more than twelve years, the Director credited the petitioner’s argument that it had not been commercialized, asserted, marked, licensed, or otherwise applied in the petitioner’s technology space. Those circumstances supported the petitioner’s own expectations and outweighed considerations favoring discretionary denial. Home Depot U.S.A., Inc. v. H2 Intellect LLC, No. IPR2025-00480, Paper 11, at 2–3 (P.T.A.B. Sept. 4, 2025) (informative Director decision). The decision illustrates an inquiry into commercial and enforcement history, rather than a determination that the patent’s technical contribution was large or small.
The evidence appropriate to that inquiry follows from its purpose. Issuance records establish the passage of time. Earlier complaints and licensing correspondence can establish the history of assertion and notice. Commercial records may show where the patent has been deployed and whether a party has organized investments or business relationships around the relevant expectations. These are potential sources of evidence, not a claim that the existing decisions require proof of a particular investment or license in every case. The Director or an APJ applying the governing policy can evaluate their dates, credibility, and legal significance using ordinary adjudicative methods. Identifying the relevant commercial field may require context, but determining that a patent has long been licensed in an industry ordinarily does not require deciding whether its claimed algorithm differs from the cited art.
The distinction can be seen in a simple hypothetical. A patent owner invokes a decade of licensing activity, while the petitioner points to a recent, unexpected assertion in a different commercial setting. Deciding how much weight to give those competing expectations concerns the parties’ conduct and the policy reasons for allowing another proceeding. Deciding whether an earlier publication discloses every claimed step concerns the technical merits. The first decision does not become more accurate merely because its adjudicator holds an engineering degree; the second may depend heavily on the ability to evaluate engineering evidence.
Settled expectations should therefore be understood as a reason offered for allocating administrative review, not as proof that older claims are substantively valid. Dabico supplies no automatic age-based immunity, and Home Depot confirms that elapsed time need not control the result. A discretionary denial on this ground does not establish that the invention is novel or nonobvious. The case for Director leadership over this policy also does not depend on endorsing every application of the criterion. Parties may dispute the proper weight of age, notice, and reliance without turning those disputes into technical patentability questions.
C Other discretionary criteria likewise turn on procedural and institutional evidence
Serial-petition practice supplies another example. General Plastic considers prior petitions, the timing of access to earlier responses and institution decisions, delays and their explanations, the Board’s resources, and its ability to meet deadlines. Its concern includes the tactical advantage of revising successive challenges after studying earlier proceedings. Gen. Plastic Indus. Co. v. Canon Kabushiki Kaisha, No. IPR2016-01357, Paper 19, at 16–18 (P.T.A.B. Sept. 6, 2017) (precedential as to § II.B.4.i). Much of this inquiry can be answered from the filing sequence and the parties’ explanations. Determining when a petitioner received an earlier response does not require deciding whether the response’s technical position was correct. Technical assistance may still be useful when evaluating the significance of new art or the reasonableness of a search, but the central concern is fair use of the review process.
Prior adjudication presents a related allocation question. In Magnolia, the Director denied institution after an unsuccessful district-court proceeding, where the petitioner had filed its administrative challenge following exclusion of its invalidity expert. The decision emphasized the opportunity already afforded to litigate, the resources spent, and the attempt to obtain another forum after the litigation loss; it did not rest on a fresh merits determination that the asserted art failed to establish unpatentability. Magnolia, No. IPR2026-00097, Paper 17, at 9–10. Assessing that procedural history requires understanding what the court decided and why the petitioner sought another proceeding. It does not require independently reproducing the technical invalidity analysis that the petitioner wished to present.
Workload considerations are still more directly administrative. The March 2025 memorandum directs attention to ex parte appeal pendency, AIA deadlines, and other demands on the Board. March 2025 Memorandum, supra, at 3. Caseloads, available adjudicative capacity, and completion times can be evaluated through institutional data. A patent’s complexity may affect a resource estimate, but the broader choice about allocating capacity among statutory responsibilities belongs to agency management.
The March 11, 2026 memorandum adds concrete economic considerations: domestic manufacturing and investment associated with accused products, domestic manufacturing of competing products made, sold, or licensed by the patent owner, and whether a petitioner sued for infringement is a small business. Memorandum from John A. Squires, Under Sec’y of Com. for Intell. Prop. & Dir., U.S. Pat. & Trademark Off., to All PTAB Users, Additional Discretionary Institution Considerations—U.S. Manufacturing and Small Business Use of AIA Proceedings 2–3 (Mar. 11, 2026). Production records, investment evidence, financial disclosures, and enterprise characteristics are ordinarily the pertinent materials. Evaluating where a product is assembled or whether a business satisfies relevant size standards does not ordinarily require determining whether the product’s underlying technology was obvious. These criteria may raise substantial questions of policy and statutory authority, but technical specialization is not what resolves those questions.
D Technical predicates should remain with specialists
The argument for this division is strongest when its limits are explicit. The March 2025 memorandum also identifies changes in law, challenge strength, expert reliance, and compelling economic, public-health, or national-security interests. March 2025 Memorandum, supra, at 2. Some aspects are legal or policy judgments; others can turn on technical evidence. Identifying an intervening precedent is a legal task, while evaluating its effect on a particular technical disclosure may require specialization. Similarly, assessing a claimed public-health need may require scientific evidence even though deciding its institutional weight is a policy judgment.
Section 325(d) illustrates the same distinction. Under Advanced Bionics, the Office first asks whether the same or substantially the same art or arguments were previously presented and, if so, whether the petitioner has shown material Office error. Advanced Bionics, LLC v. MED-EL Elektromedizinische Geräte GmbH, No. IPR2019-01469, Paper 6, at 8–10 (P.T.A.B. Feb. 13, 2020) (precedential). Identifying an earlier submission may be straightforward, but determining whether the examiner misunderstood a reference or misapplied a claim limitation can require the very technical competence that supports APJ merits review. Magnolia likewise recognizes that a substantial showing of apparent examiner error can overcome other considerations favoring denial. Magnolia, No. IPR2026-00097, Paper 17, at 8.
The Office’s published discretionary-process guidance recognizes this need by calling for consultation with personnel possessing relevant technical expertise when the parties argue the strength of the merits. U.S. Pat. & Trademark Off., Interim Director Discretionary Process § V.B, https://www.uspto.gov/patents/ptab/interim-director-discretionary-process (last visited Oct. 4, 2026). A durable division of responsibility should make that contribution explicit. Specialist APJs should resolve any necessary technical predicate and explain its basis; the official applying discretionary policy can then determine how it affects the balance. Routine scheduling, party-identity, and reliance questions do not justify transferring the substantive merits inquiry to a generalist institutional process.
Policy supervision should therefore identify which questions require a central determination and which can be resolved through delegated application. A panel deciding a routine scheduling issue need not independently reconsider the Office’s resource priorities. Conversely, a material change in the weight assigned to settled expectations should be addressed through a transparent agency policy process rather than an unexplained departure in one case. The division preserves Director responsibility for the governing standards while allowing APJs to apply them and supply the technical analysis when needed.
The statutory assignment of institution authority to the Director does not require personal resolution of every technical issue. The regulations provide that the Board institutes trial on the Director’s behalf, and the Federal Circuit has recognized the Director’s authority to delegate institution decisions. 37 C.F.R. § 42.4(a) (2026); Ethicon Endo-Surgery, Inc. v. Covidien LP, 812 F.3d 1023, 1031–33 (Fed. Cir. 2016). A division between discretionary leadership and specialist merits adjudication thus fits within a system of accountable delegation. It does not transfer the Director’s statutory office or eliminate appropriate supervision.
V The technical work required by merits review
Merits institution is preliminary, but it demands substantive analysis. An IPR petition may challenge claims under §§ 102 and 103 on the basis of patents and printed publications. 35 U.S.C. § 311(b). Deciding whether the petition presents a sufficiently persuasive challenge requires understanding the claimed invention and determining what the cited references actually disclose. A claim chart can organize the petitioner’s assertions, but its boxes do not establish the technical accuracy of the proposed mappings.
Consider a hypothetical claim requiring a component to perform a particular operation before another component receives a signal. A reference may disclose both components and both operations while placing them in a different sequence or assigning their functions differently. Whether that disclosure meets the limitation may depend on an understanding of the architecture, the relationship among the components, and the consequences of the claimed sequence. Similar vocabulary can conceal materially different technical arrangements. A decisionmaker must be able to identify those differences before determining their legal significance.
Obviousness presents an additional layer of technical judgment. Identifying known elements does not complete the analysis. The decisionmaker must evaluate the explanation for combining or modifying the references and determine whether it supports the asserted result. The Supreme Court’s flexible approach requires an articulated explanation grounded in the circumstances, while recognizing the relevance of predictable combinations and ordinary creativity. KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 418–21 (2007). Applying that approach responsibly requires enough technical understanding to distinguish a workable proposed modification from an assertion that merely restates the desired conclusion.
Expert declarations do not remove this responsibility. They make the quality of the decisionmaker’s evaluation more consequential. An APJ must examine whether the expert has explained the relevant mechanism, identified support in the reference, accounted for contrary teachings, and justified the asserted expectation of success. Technical training can help expose an assumption hidden inside an apparently straightforward legal argument. It can also help identify when a patent owner’s technical distinction has no support in the claim language or record.
These tasks explain the comparative advantage of specialist adjudication. The relevant competence combines legal judgment with the ability to interrogate technical propositions. A centralized decisionmaker with broad administrative responsibilities can obtain assistance on those propositions. A specialist panel can integrate their evaluation directly into the decision for which it is responsible.
VI APJ qualifications are aligned with the merits inquiry
Congress expressly connected APJ service with both legal and scientific competence. Section 6(a) requires APJs to possess “competent legal knowledge and scientific ability” and provides for appointment by the Secretary of Commerce in consultation with the Director. 35 U.S.C. § 6(a). The USPTO describes its APJs as legally and technically trained, with substantial patent-law experience drawn from private practice, government service, and corporate legal departments; many have also served as patent examiners or judicial clerks. U.S. Pat. & Trademark Off., Patent Trial and Appeal Board, https://www.uspto.gov/about-us/organizational-offices/patent-trial-and-appeal-board (last visited Oct. 4, 2026).
Those qualifications support an institutional judgment about who should decide technical merits. Formal education supplies a foundation for understanding scientific concepts. Patent practice develops familiarity with claim scope, disclosure, and the relationship between technical facts and legal standards. Repeated adjudication provides experience assessing competing presentations of those matters. Their combination makes specialist APJs better equipped for sustained merits review than an institution-wide process organized around the Director’s broader responsibilities.
The Board’s current paneling procedures reinforce this connection. Standard Operating Procedure 1 assigns judges technology-discipline designations and directs those assigning panels to endeavor, wherever possible, to include at least two judges whose designations match the case’s technology discipline. It also provides for pairing judges new to a jurisdiction with more experienced judges. U.S. Pat. & Trademark Off., Standard Operating Procedure 1: Assignment of Judges to Panels 4–5 (rev. 17, July 20, 2026) [hereinafter SOP 1]. This is an operational means of bringing relevant expertise to a case, subject to availability and other assignment constraints.
The argument does not require assuming that every APJ is an expert in every invention or that academic credentials guarantee sound judgment. Nor does § 6(a) prescribe a doctorate or a fixed number of years in practice. The relevant proposition is comparative: the Office has a body of adjudicators selected for combined legal and scientific competence, with procedures intended to connect their backgrounds to the matters they decide. Giving those adjudicators primary responsibility for technical merits makes practical use of that institutional design.
VII What the scholarship supports
Empirical scholarship provides qualified support for the value of technical specialization. Matthew Sipe examined patent appeals docketed during fiscal years 2015 and 2016 and compared appellate treatment of findings originating in different forums. In an adjusted comparison, PTAB findings received higher affirmance rates than district-court judicial findings; the difference was particularly pronounced for factual findings. Matthew G. Sipe, Experts, Generalists, Laypeople—and the Federal Circuit, 32 Harv. J.L. & Tech. 575, 591, 610–11 (2019).
Sipe’s findings are consistent with the proposition that specialist adjudication can contribute to reliable technical factfinding. They do not isolate qualifications as the cause of an outcome, compare APJs with the Director’s centralized institution process, or measure the current bench. Selection effects and differences in appellate review complicate the comparison, as the article recognizes. Id. at 617–21, 623, 626–27. The research strengthens the plausibility of the institutional argument without supplying a numerical prediction of the announced reform’s effect.
Criticism of APJ experience also deserves accurate treatment. Kristen Osenga’s discussion of institutional design for § 101 adjudication relied on an earlier study of experience at appointment and included the Chief Judge’s response concerning technical credentials and professional experience. Kristen Osenga, Institutional Design for Innovation: A Radical Proposal for Addressing § 101 Patent-Eligible Subject Matter, 68 Am. U. L. Rev. 1191, 1241–42 & nn.240–44 (2019). Those historical figures are not a current census of APJ qualifications. Years elapsed since law-school graduation also measure something different from scientific training, experience as a patent agent or examiner, or competence in evaluating a particular technology.
A stronger contemporary evaluation would examine relevant education, professional practice, technical assignments, and the quality of decisions. Neither an institution rate nor a raw count of years in practice answers that inquiry. The affirmative case for APJ merits responsibility rests on the match between the adjudicators’ combined skills and the work the statute requires, with empirical research serving as a useful check on that proposition.
VIII Making the return effective
Implementation should place technically appropriate APJs in charge of merits institution decisions and provide enough explanation for the parties to understand the principal grounds. A reasoned opinion should identify the claim interpretation or prior-art dispute that matters, explain the evidentiary support for the panel’s assessment, and connect that assessment to the institution threshold. The opinion need not anticipate every issue that a full trial may develop. It should show that the decisive technical questions received meaningful consideration.
The Office should also make the relationship between discretionary and merits reasoning visible. When a denial rests on policy or case-management grounds, the decision should identify that basis. When discretionary treatment depends on the strength of a technical challenge, the Office should explain the technical assessment on which it relies. Clear reasons would allow parties to distinguish a failure to satisfy the merits threshold from a discretionary choice to decline review despite a potentially substantial challenge.
Panel selection should preserve both competence and impartiality. SOP 1 already requires impartial designation and prohibits selecting panel members to influence an outcome. It also limits the Director’s involvement in assigning or reassigning specific proceedings before a panel decision. SOP 1, supra, at 1–2. Matching technology backgrounds to a case should operate within those protections. Relevant experience is a reason for assignment; a predicted disposition is not an appropriate substitute.
The implementing directive should distinguish policy ownership, case-specific application, and supervisory review. It should identify which discretionary matters remain reserved to the Director, which APJs may resolve under established policy, and how novel questions or significant departures reach the responsible official. It should also specify how panel members deliberate, who is responsible for the written reasons, and how any supervisory revision of the merits analysis will be explained. These features would make direct panel responsibility a substantive procedural arrangement rather than a change in the name appearing on an institution notice.
Other statutory institution requirements require their own allocation. The directive should identify who resolves issues such as § 315(b), how the parties present the relevant evidence, and the procedure for any available review. Mandatory restrictions must be applied according to their governing law even when their underlying facts are nontechnical. The effective date and treatment of pending petitions should be equally clear.
Evaluation should focus on timeliness, the clarity and consistency of reasoning, the treatment of technical disagreements, and whether supervisory intervention improves decisions without unnecessary duplication. Those measures should also inform comparison with a system of published APJ recommendations and explained Director departures. Institution percentages remain useful descriptive information, but a higher or lower rate does not by itself establish that the technical analysis has improved.
IX Whether the institution panel should also conduct the trial
Returning merits institution determinations to the PTAB leaves a consequential question about who will handle the proceedings after institution. The announcement itself does not resolve whether the same APJs will conduct the trial and issue the final written decision. That choice deserves express attention because restoring specialist responsibility and preserving the same individuals throughout a case address different institutional objectives.
Continuity has substantial advantages. The institution panel has already studied the patent, the asserted references, and the principal disputes. Keeping that panel avoids requiring another group to repeat the initial work and allows the trial to build on an established understanding of the technology. That familiarity can assist case management and sharpen the issues on which additional evidence would be useful. Replacing the panel consumes resources and may sacrifice some of those benefits.
Continuity also creates a plausible risk of anchoring. An initial understanding of a claim, reference, or proposed combination may become the point from which the judges evaluate later submissions. Even when the preliminary view was reasonable on the available record, it may exert too much influence after that record changes. The concern is insufficient adjustment: a panel might carry forward an early technical premise despite testimony or cross-examination that materially weakens it.
Judicial expertise does not eliminate the need to consider this possibility. Experimental research has shown that numerical anchors can influence judges’ hypothetical decisions. Chris Guthrie, Jeffrey J. Rachlinski & Andrew J. Wistrich, Inside the Judicial Mind, 86 Cornell L. Rev. 777, 790–93 (2001). Applying that insight to a preliminary technical opinion is an analogy, rather than a demonstrated numerical-anchoring effect in PTAB cases. The researchers themselves caution that results from hypothetical problems do not conclusively establish error in actual adjudication. Id. at 819–21. The evidence supplies a reason to examine procedural design, rather than a basis for presuming that an individual panel will decide unfairly.
A related concern is confirmation bias. Once a panel has explained why a petition presents a sufficiently strong challenge to justify trial, the judges may unconsciously favor later evidence consistent with that explanation. Contrary evidence may receive more skeptical treatment, or an ambiguity may be resolved in a way that preserves an earlier interpretation. Scholarship specifically addressing IPR has advanced this concern as a reason to separate institution and final adjudication. See, e.g., Adam J. Cook, Comment, The Same PTAB Panel Should Not Do It All: Why Inter Partes Review Decisions Should Be Bifurcated, 128 Penn St. L. Rev. 257, 280–81 (2023). That argument is a proposed institutional reform, not an empirical finding that current APJs necessarily exhibit such bias.
The legal distinction between institution and final adjudication should reduce the perceived need to defend an earlier conclusion. Institution asks whether the preliminary showing crosses a statutory threshold. At trial, the petitioner must prove unpatentability by a preponderance of the evidence, and the Board must issue a final written decision if the instituted review is not dismissed. 35 U.S.C. §§ 316(e), 318(a). A final decision that the petitioner has failed to carry its burden does not necessarily show that institution was mistaken. A reasonable initial prediction can be displaced by a more complete record.
The Federal Circuit has made the need for that fresh assessment explicit. In TriVascular, it explained that the Board is not bound by its preliminary institution findings and should change its view when the developed record warrants doing so. TriVascular, Inc. v. Samuels, 812 F.3d 1056, 1068 (Fed. Cir. 2016). In Magnum Oil Tools, it rejected an approach that shifted the burden to the patent owner based on institution and required consideration of the merits anew on the total record. In re Magnum Oil Tools Int’l, Ltd., 829 F.3d 1364, 1375–77 (Fed. Cir. 2016). Institution therefore cannot become a presumption that the challenged claims are unpatentable.
Existing precedent also prevents treating panel continuity as inherently unlawful. In Ethicon, the Federal Circuit rejected statutory and due-process objections to the same panel deciding institution and final patentability, where no additional procedural-fairness infirmities were alleged. Ethicon, 812 F.3d at 1029–31. The policy question remains whether a different arrangement would improve adjudication or public confidence. Legal permissibility does not establish that every permissible design is equally effective.
The Office could preserve continuity, use separate technically qualified panels, or introduce partial rotation. Separate panels could reduce the personal commitment associated with defending one’s own earlier reasoning while preserving specialized adjudication at both stages. They would still encounter the institution opinion, however, so reassignment would not necessarily eliminate anchoring. Partial rotation might retain technical familiarity while introducing additional perspectives, but its benefits likewise would need evaluation.
Whichever approach the Office selects, the final decision should directly address material evidence developed after institution and explain why the preliminary technical assessment remains persuasive or should change. Deliberation should include express consideration of the strongest contrary evidence. Transparent assignment practices and a clear expectation of reconsideration would reinforce the principle that expertise serves the record as it develops. The return to APJ merits responsibility offers an opportunity to combine technical competence with procedures that encourage that openness.
X A durable allocation of responsibility
Returning ordinary merits institution decisions to PTAB panels is best understood as a proposal about responsibility for deliberation and reasons. The centralized process already has access to APJ expertise. Direct panel adjudication would make collective technical judgment and a reasoned explanation attributable to the deciding specialists the ordinary basis for the merits determination. Its advantage depends on those practices and on effective supervision, not merely on the identity of the official issuing the decision. A system of published APJ recommendations and explained Director departures could achieve many of the same benefits; direct panel decisions offer the simpler default when another routine approval step adds little.
The Director’s comparative advantage lies in establishing and supervising lawful Office-wide discretionary policy, including how the agency weighs competing institutional interests and allocates resources. That responsibility permits delegation of routine applications to APJs. Trial schedules, litigation investment, party identity, and the history supporting settled expectations can ordinarily be evaluated without deciding how the claimed invention operates. Their nontechnical character does not alone establish that the Director should personally decide them, and it does not turn mandatory institution restrictions into discretionary considerations.
The return should be accompanied by reasoned opinions, impartial assignments informed by technical experience, defined supervisory procedures, and an express decision about panel continuity. A system that gives specialists responsibility for technical merits must also support their willingness to reconsider preliminary views. With those features, the Office can preserve accountable policy leadership while improving the conditions for careful and open-minded patentability adjudication.
