Patent Invalidity, Patentability

Obviousness rejections are among the most, if not the most, frustrating rejections a patent applicant can face. The examiner typically assembles two or three old patents or published patent applications as prior art references, announces that a person of ordinary skill in the art (a “POSITA”) would have combined them, and then supplies a reason why. Since the Supreme Court decided KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398 (2007), that reason no longer has to come from the patent or from the references themselves. Market forces, design incentives, and “universal” motivations known in a field will all do the job. What is easy to overlook is that there is a gate the reference has to pass through before any of that matters. The reference has to be “analogous art” in the first place. A decision the Federal Circuit handed down on August 14, 2026 suggests that this gate is opening wider – and that the breadth of your own claims may be what opens it.

The analogous art test comes from In re Bigio, 381 F.3d 1320, 1325 (Fed. Cir. 2004), and it has two prongs. A reference qualifies if it is from the same “field of endeavor” as the claimed invention. If it is not, it can still qualify if it is “reasonably pertinent to the particular problem with which the inventor is involved.” The second prong asks whether the reference “logically would have commended itself to an inventor’s attention in considering his problem.” In re Clay, 966 F.2d 656, 659 (Fed. Cir. 1992). Everything therefore turns on how the problem is defined, and in practice the place everyone has looked for the problem is the patent itself – usually the Background section, where the applicant recites the shortcomings of what came before. That has been a modest but real defensive tool. If the patent says the problem is X, a reference addressed to Y is arguably outside the field of view of a POSITA working on X. 

The analogous arts test acts as a check and balance on examiners citing prior art references that no one confronted with the problem the inventor was facing would have ever consulted, and it acknowledges the fact that invention always has a context in which it occurs. 

That check and balance has potentially lost its potency. In Nielsen Company (US), LLC v. TVision Insights, Inc., No. 2025-1371 (Fed. Cir. Aug. 14, 2026), the Court affirmed a Patent Trial and Appeal Board decision cancelling claims of Nielsen’s audience measurement patent. The patent covered a system that uses a camera to capture images of the people watching a television, reduces the resolution of an image to determine the orientation of a viewer’s head, and then uses higher resolution images to run facial recognition. The Background of the patent identified a specific problem: the illumination source used to light the room was being activated too often, which drained power, generated heat, shortened the life of the bulb, and annoyed the audience. The prior art reference at issue, a scientific paper titled “Evaluation of Face Resolution for Expression Analysis,” said nothing about illumination sources or audience measurement. It studied how well facial expression analysis performs on down-sampled, lower resolution images. A copy of the slip opinion can be obtained here: 

Nielsen argued that the problem confronting its inventor was the one its specification expressly described and that the Board’s analysis had to be “tethered to” what the patent identified as the problem facing the inventor. The Federal Circuit disagreed. First, relying on KSR itself, the Court reiterated that “[t]he problem motivating the patentee may be only one of many addressed by the patent’s subject matter,” and that a reference is analogous if it is reasonably pertinent to one or more of the problems to which the patent relates. See Donner Tech., LLC v. Pro Stage Gear, LLC, 979 F.3d 1353, 1359 (Fed. Cir. 2020). None of that is new.

What is new is where the Court looked to find the additional problems. It looked at the claims. “It would be inappropriate to limit the relevant problem to reducing light-source use,” the Court held, “when the claims themselves are not limited to the presence of a light source and are far broader than responding to the light-source problems Nielsen identifies.” And for that proposition the Court quoted perhaps the most quotable line in KSR: “In determining whether the subject matter of a patent claim is obvious, neither the particular motivation nor the avowed purpose of the patentee controls. What matters is the objective reach of the claim.” 550 U.S. at 419.

Why does that matter? Because KSR said that about the motivation to combine, which is the step that comes after the analogous art threshold has been cleared. It has long been understood that the reason for combining references need not appear anywhere in the patent under attack. The threshold question of whether a given reference was even relevant enough for a POSITA to consult, by contrast, has generally been anchored to the subject patent’s own account of the problem being solved. In Nielsen, the objective reach of the claim now supplies the anchor. If the claims are broad enough to be read as addressing a general problem – here, image processing and facial detection – then art addressed to that general problem is fair game, even though the problem appears nowhere in the specification. This means, however, that if the claims address an unstated problem addressed by the prior art, that prior art is now citable in an obviousness rejection. The problem is that under this decision, the reference would appear to be citable against a narrower, dependent claim that does not address that same unstated problem. 

In fairness, Nielsen does not appear to be a case in which the “problem” upon which the Court relied was untethered to the problems described in the patent. The Court went on to say that the Board did not need to look beyond the patent at all, because the title (“Methods and Apparatus to Capture Images”), the claims, and the Background section together showed that the patent was concerned with image processing and facial recognition. But the claim-scope rationale is stated as an independent ground, and it is the part of the opinion that petitioners and examiners are going to cite. The practical effect is that the pool of assertable references is defined less by what the applicant said the invention was for and more by how much ground the claims actually cover.

This case is another cautionary tale about the hazards of claiming an invention too broadly. Inventors tend to be very focused on getting maximum claim breadth to prevent infringers from designing around their patents. There has always been a risk that if you claim too broadly, you may run into the prior art. That has always been the nature of patents, and dependent claims have been the chief way to mitigate this risk. However, Nielsen heightens this problem by tying claim breadth to the scope of what prior art references can be considered in the first place, and it does not appear that narrow dependent claims will help. Regarding the treatment of dependent claims in Nielsen, only two dependent claims of the Nielsen patent actually recited an “illumination source” – the very element tied to the problem the specification described – and the petitioner did not challenge either of them. So, it is unclear whether the Court would have applied the analogous art test differently to those claims.