“The PTAB remains a patent death squad for those patent owners who are unlucky enough to be trapped in instituted proceedings from the old regime. The statistics prove it.”
It is not news to anyone in the industry that the Patent Trial and Appeal Board (PTAB) has changed. Thanks to U.S. Patent and Trademark Office (USPTO) leadership, discretionary denials have increased, institution rates have dramatically dropped, and that means patent owners are finally seeing relief at the front end of the inter partes review (IPR) process. There is great reason for the justifiable optimism you hear from patent owners.
But not so fast! That the PTAB is heading in the right direction does not mean the PTAB has been fixed.
The institution rate may have fallen from 74% in FY 2024 to 43% so far in FY 2026, which is unquestionably a positive development for patent owners. But once an IPR is instituted, the patent owner remains in grave danger. The PTAB’s post-institution kill rate has not meaningfully changed.
During FY 2024, of those IPRs and post grant review (PGRs) reaching a final written decision, 312 found all claims unpatentable, 67 were a mixed ruling, and 67 found all claims patentable. So that is 70% all unpatentable, with 85% finding at least one claim unpatentable (see slide 10).
During FY 2025, of those reaching a final written decision, 250 found all claims unpatentable, 75 were mixed, and 62 found all claims patentable. So, that is 64.6% all unpatentable with 84% finding at least one claim unpatentable (see slide 11).
From October 1, 2025 through March 31, 2026—the last period for which the USPTO has made this data available—of those reaching a final written decision, 149 found all claims unpatentable, 46 were mixed, and 49 found all claims patentable. So, that equates to 61% all unpatentable with 80% finding at least one claim unpatentable.
So, yes, fewer patents are being dragged into full-blown PTAB trials, which is good. But for those patent owners who still find themselves in an instituted IPR, the PTAB remains what it has always been: a patent death squad.
The Real Problem Is Not Yet Behind Us
Why has the kill rate remained essentially unchanged despite a more patent-owner-friendly policy on institution? The answer is straightforward: the reforms have not operated retroactively.
An institution decision has always involved two overlapping inquiries. First, there is a discretionary inquiry: should the Office institute review at all? Second, there is a merits inquiry: has the petitioner shown a reasonable likelihood that at least one challenged claim is unpatentable?
Historically, those questions were bundled together and decided by the same PTAB panel. A panel of Administrative Patent Judges (APJs) that would ultimately decide the fate of the patent would first decide whether there was any reason the Office should not conduct an IPR proceeding. If the petitioner cleared this discretionary consideration, then the panel would determine if the petition cleared the “reasonable likelihood” threshold to institute on the merits. If the petition did satisfy the “reasonable likelihood” threshold, the same APJs would then preside over the trial and ultimately decide whether they were right the first time—which surprise, surprise, they nearly always found they were right all along.
Obviously, the former process was not a neutral procedure. It was an institutional design that invited confirmation bias.
USPTO Director John Squires recognized this problem in his October 17, 2025, Open Letter, explaining that panels were effectively “filling their own dockets.” He was exactly right. When the same judges who institute review later decide the final merits, the process creates a natural and powerful incentive to validate the earlier institution decision. No judge likes to admit that a case should not have been instituted. No panel wants to confess that it overlooked a fatal flaw in the petition. And no tribunal is immune from the basic human tendency to defend its own prior reasoning.
To his credit, Director Squires moved quickly to separate the APJs who assist in deciding institution from the APJs who ultimately decide patentability after trial. That was an important reform, but it has not solved the problem for cases already in the pipeline.
At this time, the same APJs who instituted IPRs under the old regime are still issuing final written decisions in those very same cases. And too often—in fact, very often—those final written decisions simply confirm the institution decision. And that is at the heart of the problem, and why the PTAB is still operating like the death squad of old. The PTAB may be improving prospectively, but patent owners are still being injured by the old system.
Confirmation Bias Showing Up in Outcomes
The defenders of the PTAB will say talk of confirmation bias is merely rhetorical and anecdotal. They will say the Board is just applying the law. They will also say patent owners are losing because the challenged claims are weak. But the numbers tell a different story.
At the same time the Office congratulates itself for examiner quality, are we really supposed to simultaneously believe that in roughly 70% of instituted IPRs every challenged claim is invalid? And this is just with respect to the patents that are valuable enough to litigate. If examiners are completely wrong on 70% of patents valuable enough to litigate, then how wrong are they across the board when we all know that upwards of 90% of patents wind up not being valuable at all? If examiners are accurate more than 96% of the time, as we are told year after year, how is it they are so wrong on so many patents that are valuable enough to litigate? And by litigate we are talking about spending $500,000 to $1,000,000 just in legal fees, just at the PTAB.
It seems entirely fantastical to believe that litigated patents valuable enough to be asserted, defended, and challenged are so overwhelmingly defective that the PTAB is justified in wiping out all challenged claims in case after case. And when you add in the obvious intellectual disconnect with examiners receiving A+ grades each year, something does not compute.
What is far more plausible is that once the PTAB institutes review under what is a truly very low threshold, the deck is stacked against the patent owner. If a panel has already said there is a reasonable likelihood that at least one claim is unpatentable, and then that same panel applies a preponderance of the evidence standard, the path to cancellation is not difficult to predict. The panel has already previewed its view of the case. The final written decision then becomes less of an independent adjudication and more of a validation exercise—and that is precisely a confirmation bias.
More Than Prospective Reform Needed
The good news is that the USPTO has finally recognized some of the structural problems that have plagued the PTAB since the America Invents Act (AIA). Increasing discretionary denials and separating institution-stage APJs from merits-stage APJs are meaningful reforms. But patent owners should not be expected to absorb the damage caused by the old process simply because reform arrived too late. Indeed, if the old process was flawed to the point that it needed to be fundamentally reset, it should be self-evident that those cases in process that were instituted using that flawed former process demand new scrutiny.
And this is not a one-off problem here and there. There remains a substantial inventory of cases in which the same APJs who instituted review are still deciding final patentability. If the point of the reform is to eliminate bias and stop the weaponization of the PTAB, then final written decisions emerging from the pre-reform structure should not be treated as business as usual. The Director has the authority to do something about this, and these legacy cases demand a fresh look from the Director.
When the constitutionality of the PTAB was challenged in United States v. Arthrexthe Supreme Court made clear that the Director must have ultimate review authority over PTAB decisions. That authority exists for a reason. It is the constitutional safety valve that allows the politically accountable head of the Office to correct agency adjudication when the PTAB goes off course. A fresh look at the legacy cases is precisely the type of situation where Director review matters and should be undertaken.
If a final written decision issues from a panel that also instituted the proceeding, and the final decision merely ratifies institution despite serious petition defects, the Director should take a hard look. For example, cases like Motorola Mobility LLC v. Multifold Int’l Inc. Pte. Ltd., IPR2025-00040 (petition fails to specify where each element of the claim was found in the prior art, as required by 37 C.F.R. § 42.104(b)(4)) and Apple Inc. v. DH Int’l Ltd., IPR2025-00172 (a Japanese-language foreign document presented with a declaration signed by an individual who lacked personal knowledge of the translation) would never be instituted under the current process. And if that is true, why should the patent claims be invalidated now by a final written decision?
Simply stated: if the Board supplied missing petition arguments, excused facial defects, overlooked translation failures, or otherwise rescued a deficient petition, the Director should intervene.
The PTAB Is Better, But It Is Not Fixed
Patent owners have reason to be encouraged by recent developments at the USPTO, but the job is not done.
The PTAB remains a patent death squad for those patent owners who are unlucky enough to be trapped in instituted proceedings from the old regime. The statistics prove it. The structure explains it. Recent final written decisions confirm it.
The Director should not wait for the new system to work its way through the pipeline. Patent owners are still losing property rights today because of a flawed legacy process the Office has already recognized as unfair and required fundamental change. So, while prospective reform is welcome it is not enough. Because of the legacy problems at the PTAB, patent owners continue to pay the price for a system that continues to be tilted against them.
Image Source: Deposit Photos
Author: ymgerman
Image ID: 71962199