Patent law rarely changes all at once. It moves the way a glacier moves: imperceptibly day to day, and then suddenly you notice the landscape is different. The second half of 2024 was one of those moments when the change became visible. Within a few months, the Federal Circuit threw out a fifty-year-old framework for judging the obviousness of design patents, the dust from a unanimous Supreme Court decision on enablement was still settling across the life-sciences industry, the Supreme Court resolved a closely watched intellectual-property case on the strength of "history and tradition" rather than doctrine, the question of what software can even be patented remained as unsettled as it had been a decade earlier, and the agency that issues patents kept rewriting the rules for challenging them after they issue.

This article is a strategic field guide to that period and the trajectory it set. It is written for three audiences at once. If you are a judge or a lawyer, you will find the holdings, the statutory hooks, and the citations you expect. If you are an inventor, a founder, or an in-house product lead who has never read a patent opinion in your life, you will find every term of art explained in plain language the first time it appears, with concrete examples. And if you are an IP portfolio owner trying to decide where to spend your filing budget next quarter, you will find, for each development, a candid answer to the only question that ultimately matters: what should I do differently because of this?

A word of honesty up front. Some of these matters were still pending when the original version of this piece was written in 2024. Courts decide cases on their own schedule, and several of the disputes discussed below have since produced rulings, denials of certiorari, or further proceedings on remand. Rather than pretend otherwise, we treat each topic as a trajectory — a direction the law is heading — and we flag where the destination is now known versus where it remains genuinely open. That framing is more durable than a box score, and it is more useful for planning a patent portfolio that has to survive twenty years.

A One-Paragraph Map of the Terrain

Before diving in, here is the shape of the whole. Seven forces were reshaping patent and adjacent IP strategy heading out of 2024. First, design-patent obviousness was overhauled by the en banc Federal Circuit in LKQ v. GM, replacing a rigid two-prong test with the same flexible standard used for ordinary inventions. Second, the design-patent damages regime under 35 U.S.C. § 289 — the engine that makes design patents so valuable, and so dangerous to infringers — continued to be worked out in the wake of Samsung v. Apple. Third, enablement — the rule that you must teach the public how to make and use what you claim — was tightened for broad genus claims by the Supreme Court in Amgen v. Sanofi. Fourth, patent eligibility under Section 101 remained mired in the Alice/Mayo framework, with no Supreme Court rescue and persistent unpredictability for software and diagnostics. Fifth, in Vidal v. Elster the Supreme Court showed how it now decides IP cases that touch the Constitution — through history and tradition — a tell that matters for anyone hoping the Court will rescue them from a doctrine they dislike. Sixth, the USPTO's examination guidance kept evolving, most notably with updated eligibility guidance addressing artificial intelligence. Seventh, the Patent Trial and Appeal Board (PTAB) and the inter partes review (IPR) system continued to swing on the pendulum of discretionary denials, the Fintiv saga lurching between expansion and restriction depending on who led the agency — and, throughout, patent damages continued their long migration toward strict apportionment. We take each in turn.

Part One: The Earthquake in Design Patents — LKQ Corp. v. GM Global Technology Operations

What a design patent is, in plain terms

Most people picture a patent as protecting how a machine works — its function, its mechanism, its method. That is a utility patent. A design patent protects something different: how an article of manufacture looks. The ornamental shape of a chair, the contour of a car fender, the icon arrangement on a phone screen, the silhouette of a beverage bottle — these are the province of design patents under 35 U.S.C. § 171. A design patent has no written claim describing the invention in words; the claim is the drawings. If you have ever seen a patent filing that is just a series of line drawings of a sneaker from six angles, you have seen a design patent.

Design patents have quietly become enormously valuable. They are cheaper and faster to obtain than utility patents, they last fifteen years from grant, and — as anyone who followed the Apple v. Samsung smartphone wars learned — they can support staggering damages awards because of a unique statutory remedy (35 U.S.C. § 289) that lets the patent owner recover the infringer's total profit on the article. We unpack that remedy in detail below. For a broader treatment of how design and utility protection interact, see our companion article on the intersection of design and utility patents, and for the related question of how design patents compare with trade dress, our analysis of design patents versus trade dress protection for product configurations.

The issue: an obviousness test built for design patents alone

To get any patent, an invention must be non-obvious — meaning it cannot be a trivial, predictable variation on what already existed (the "prior art") that any ordinarily skilled person would have arrived at. For utility patents, the governing framework comes from the Supreme Court's decision in Graham v. John Deere Co., 383 U.S. 1 (1966), and was made deliberately flexible by KSR International Co. v. Teleflex Inc., 550 U.S. 398 (2007). KSR rejected rigid formulas and instructed courts to apply "an expansive and flexible approach" grounded in common sense. Our comprehensive guide to Section 103 obviousness analysis walks through that framework in detail.

Design patents, however, had long been judged by a separate, far more rigid test. Under In re Rosen, 673 F.2d 388 (C.C.P.A. 1982), and Durling v. Spectrum Furniture Co., 101 F.3d 100 (Fed. Cir. 1996), an examiner or a challenger had to clear two demanding hurdles. First, they had to identify a single prior-art reference — a "Rosen reference" — whose design characteristics were "basically the same" as the claimed design as a whole. If no single prior design was close enough to qualify as a Rosen reference, the analysis ended there and the design was non-obvious, full stop. Second, if a Rosen reference existed, any secondary references used to fill in differences had to be "so related" to the primary reference that features in one would suggest application to the other. This was the Rosen-Durling test, and it was notoriously protective of design patents: by demanding a near-identical single starting point, it made many designs effectively immune from obviousness challenges.

The tension was obvious. The Supreme Court in KSR had condemned rigid, formulaic obviousness tests for utility patents — yet design patents continued to live under exactly such a test. Something had to give.

The status: the en banc Federal Circuit overrules Rosen-Durling

In LKQ Corp. v. GM Global Technology Operations LLC, 102 F.4th 1280 (Fed. Cir. 2024) (en banc), the full Federal Circuit gave way. The case arose from an IPR in which LKQ — a supplier of aftermarket auto parts — challenged a GM design patent on a vehicle fender. The PTAB, applying Rosen-Durling, upheld the patent because LKQ could not identify a Rosen reference basically the same as GM's design. The en banc court used the case to dismantle the framework.

The court held that the rigid Rosen-Durling requirements "are improperly rigid" and cannot be squared with KSR and, behind it, Graham. It overruled the two-prong test and held that design-patent obviousness must be analyzed under the same flexible Graham factors that govern utility patents: (1) the scope and content of the prior art; (2) the differences between the prior art and the claimed design; (3) the level of ordinary skill in the pertinent art; and (4) objective indicia of non-obviousness such as commercial success, industry praise, and copying. Critically, the court eliminated the threshold "basically the same" Rosen requirement — a challenger no longer needs a single near-identical primary reference to even get in the door — though it acknowledged that an obviousness analysis must still begin with a primary reference, just one that need only be "something in existence" with characteristics that the designer would have looked to, not a near-twin of the claimed design.

The aftermath confirms the seismic framing. The Supreme Court denied certiorari, leaving LKQ the law of the land, and the case returned to the PTAB on remand to re-run the obviousness analysis under the new, more permissive standard. The USPTO moved quickly to bring examination into line, issuing an updated examiner guidance memorandum in the fall of 2024 directing examiners to apply the Graham factors to design applications and to abandon the rigid Rosen threshold — which means the practical effect of LKQ is felt not only in litigation but at the examiner's desk, where design applications now face a meaningfully harder obviousness inquiry on the way in.

We have a dedicated deep dive on the decision and its mechanics in The LKQ Decision: A Seismic Shift in Design Patent Obviousness Analysis. For present purposes, the strategic significance is what matters.

A worked example

Suppose, hypothetically, that a kitchen-appliance maker holds a design patent on a stand mixer with a distinctive rounded head and a sculpted, tapered base. Under Rosen-Durling, a competitor wanting to invalidate that patent in an IPR had to find a single prior mixer that already looked "basically the same" — same rounded head, same tapered base, same overall visual impression. If the closest prior mixer had a boxy head, the competitor was often stuck at the threshold, and the patent survived without the merits ever being reached.

After LKQ, the competitor's path opens up. It can now start with a prior mixer that has the tapered base, point to a separate prior appliance (say, a blender) with the rounded head, and argue that an ordinary designer of kitchen appliances — familiar with the whole field — would naturally have combined those readily available design features. The patent owner's best response is no longer "you don't have a Rosen reference," because that defense is gone. Instead, the owner must marshal the fourth Graham factor: evidence that the mixer's look won design awards, that reviewers praised its appearance, that competitors copied it, and that its sales were driven by its distinctive form. The center of gravity in design-patent disputes has moved from a threshold gatekeeping question to a full-blown factual fight over obviousness — exactly as it has long been for utility patents.

The IP-strategy implication

LKQ made design patents easier to invalidate and harder to obtain. That single sentence reverberates across an entire portfolio strategy.

For patent owners and prosecutors. The era in which a design patent was nearly bulletproof against obviousness is over. Going forward, expect examiners and PTAB panels to combine multiple, less-similar references the way they do in utility cases. The practical defense is to build a record of objective indicia — the fourth Graham factor — early and deliberately. If a product design wins industry design awards, generates press coverage praising its look, is copied by competitors, or drives commercial success specifically attributable to its appearance, that evidence becomes a frontline defense to obviousness. Smart owners now document this evidence contemporaneously rather than scrambling for it during litigation. Prosecutors should also consider filing more design patents covering variations and partial designs, so that a portfolio presents a thicket of overlapping claims rather than a single vulnerable patent.

For accused infringers and challengers. LKQ is a gift. A defendant facing a design-patent assertion now has a realistic obviousness defense even where no single prior design matches the patent. The work shifts to assembling a coherent narrative from a constellation of prior designs and explaining why an ordinary designer would have combined them — exactly the kind of argument KSR invited in the utility context.

For valuation and licensing. Because design patents are now somewhat weaker as a class, their settlement and licensing value should be recalibrated. A design-patent demand that would have commanded a premium under Rosen-Durling now carries litigation risk that did not exist before. This affects how both sides price a patent license.

The trajectory here is settled: Rosen-Durling is gone, certiorari was denied, and the lower tribunals are now working through the inevitable second-generation questions — how analogous a primary reference must be, how the "ordinary designer" is defined, and how heavily objective indicia weigh. Those questions will be litigated for years, but the direction is fixed.

Part Two: Why Design Patents Hit So Hard — The § 289 Total-Profit Remedy

A development in invalidity law like LKQ only matters because the underlying right is worth so much when valid. Design patents punch far above their weight on damages, and understanding why is essential to weighing what LKQ really changed.

The article-of-manufacture problem

Section 289 of the Patent Act provides that anyone who applies a patented design to "any article of manufacture" is liable to the patent owner "to the extent of his total profit." Read literally, that is breathtaking: it lets a design-patent owner recover the infringer's entire profit on the infringing article, with no apportionment to the design's contribution. There is no equivalent in utility-patent law, where, as we discuss below, apportionment is the dominating theme.

The fight, naturally, has been over what counts as the relevant "article of manufacture." In the Apple v. Samsung smartphone wars, a jury awarded Apple roughly $399 million — Samsung's entire profit on the infringing phones — for infringement of Apple's design patents covering the rounded-rectangle face, the bezel, and the grid of colorful icons. Samsung's objection was intuitive: should it really forfeit all the profit on a sophisticated smartphone because three ornamental features were copied?

In Samsung Electronics Co. v. Apple Inc., 580 U.S. 53 (2016) (137 S. Ct. 429), a unanimous Supreme Court agreed that the "article of manufacture" need not be the entire end product sold to consumers; it can be a component of that product. The relevant article is the thing to which the design is applied, which may be the whole product or merely a part of it. The Court declined, however, to announce a test for identifying the article in a multicomponent product — leaving that to the lower courts, where it remains a recurring, fact-intensive battleground.

Functionality is the gatekeeper

A second limit on design-patent power is the requirement that the protected design be ornamental, not dictated by function. A design that is purely functional cannot be protected by a design patent, and the Federal Circuit has applied that principle even to spare auto-body parts whose shapes are dictated by the need to match the rest of the vehicle. In Automotive Body Parts Ass'n v. Ford Global Technologies, LLC, 930 F.3d 1314 (Fed. Cir. 2019), the court rejected an argument that Ford's design patents on replacement hoods and headlamps were invalid as functional merely because consumers wanted parts that matched their cars — aesthetic appeal, even aesthetic appeal driven by a desire to match, is still ornamental, not functional. The functionality line is subtle and consequential, and we explore it at length in The Evolution of Design Patent Functionality: Clarifying the Muddle and in our analysis of the intersection of design and trade dress.

The IP-strategy implication

The combined lesson of LKQ and § 289 is that design patents became, in the span of a few years, both easier to invalidate and, where valid, no less devastating in their remedy. That is a sharper risk profile, not a diminished one.

For owners, the play is to obtain design patents that are (a) genuinely non-obvious under the new Graham standard, supported by contemporaneous objective indicia, and (b) drawn to claim the design as applied to the whole salable product where the facts support it, since the breadth of the "article of manufacture" drives the § 289 recovery. Partial-design claiming is a powerful tool, but it can also narrow the article and the available profit base; the choice should be deliberate.

For accused infringers, § 289 is the reason a seemingly modest design-patent suit can carry company-threatening exposure. The defense agenda runs on two tracks: invalidate under LKQ, and — if the patent survives — fight hard at trial to define the "article of manufacture" as a small component rather than the whole product, dragging the total-profit base down toward something proportionate. This interacts directly with patent infringement litigation strategy and with the firm's broader design-patent litigation playbook.

Part Three: The Enablement Reckoning — Amgen Inc. v. Sanofi

What enablement means, in plain terms

The patent bargain is simple in concept: in exchange for a time-limited monopoly, the inventor must teach the public how to make and use the invention, so that when the patent expires anyone can practice it. That teaching obligation lives in 35 U.S.C. § 112(a) and is called the enablement requirement. The legal test, settled for decades, is whether a person of ordinary skill in the art (a "POSITA" — the patent law's reasonable, technically competent reader) could make and use the full scope of the claimed invention without undue experimentation. Courts weigh the In re Wands factors, 858 F.2d 731, 737 (Fed. Cir. 1988), which include the quantity of experimentation needed, the amount of guidance in the specification, the presence of working examples, the predictability of the field, and the breadth of the claims.

The phrase that does all the work is "full scope." If you claim broadly, you must enable broadly.

The issue: claiming a function and a whole genus of molecules

Amgen concerned cholesterol drugs. Amgen held patents that claimed not a specific antibody by its chemical structure, but an entire genus of antibodies defined by what they do: any antibody that binds to particular amino-acid residues on the PCSK9 protein and thereby blocks PCSK9 from degrading the body's LDL-receptors. By some estimates the claims covered millions of possible antibodies. Amgen's specification disclosed the amino-acid sequences of twenty-six examples and offered two general methods — a "roadmap" and a process of "conservative substitution" — for finding others.

Sanofi, which marketed a competing PCSK9 antibody, argued that this was a classic over-claim: Amgen had claimed an entire functional universe of molecules while disclosing only a tiny sample and a research program for discovering the rest. To practice the full scope, a scientist would have to engage in exactly the kind of trial-and-error screening that the law calls undue experimentation.

The status: a unanimous Supreme Court tightens the screws

In Amgen Inc. v. Sanofi, 598 U.S. 594 (2023), the Supreme Court unanimously agreed with Sanofi and affirmed the invalidation of the broad claims. Justice Gorsuch's opinion reaffirmed a principle running back to the nineteenth-century cases of O'Reilly v. Morse (the telegraph) and The Incandescent Lamp Patent (Edison's filament): "the more a party claims, the broader the monopoly it demands, the more it must enable." A patentee may not claim an entire functional genus while leaving scientists to discover most of its members through their own painstaking experimentation. Disclosing examples and a general method for finding the rest, the Court said, "is not enabling more than a quantitative subset of the claimed embodiments — it is instructing scientists to engage in painstaking experimentation to see what works."

Importantly, the Court did not hold that functional or genus claims are categorically invalid, and it did not impose a "reasonable time and effort" gloss or a new test. It applied the established full-scope standard. But the message was unmistakable: broad claims that recite a result and outsource the work of getting there to the reader are vulnerable.

This builds on a line the Federal Circuit had already been developing in the Amgen litigation itself (see Amgen Inc. v. Sanofi, 872 F.3d 1367, 1373 (Fed. Cir. 2017)) and in cases like Wyeth & Cordis Corp. v. Abbott Laboratories, 720 F.3d 1380 (Fed. Cir. 2013), which invalidated functional pharmaceutical claims that would have required synthesizing and screening "millions of compounds." The USPTO's own examination practice reflects the same principle: examiners issue § 112(a) enablement and written-description rejections, particularly in "unpredictable technologies" like chemistry and biotechnology, precisely where the gap between a broad functional claim and what the specification actually teaches is widest. (See MPEP §§ 2164, 2164.01(a).) The aftermath of Amgen in the lower courts has been a steady stream of full-scope enablement invalidations of broad functional and genus claims — the decision did not fade; it became a workhorse defense.

The IP-strategy implication

Amgen reshaped how sophisticated patentees draft and how challengers attack — most acutely in life sciences, but the logic reaches any field that uses functional or genus claiming.

Draft a claim ladder, not a single broad claim. The durable response is to claim in a graduated structure: a broad independent claim, then progressively narrower dependent claims that recite specific structures, sequences, species, or working examples. If the broad claim falls to an enablement attack, the narrower claims — anchored to what the specification actually teaches — survive. A portfolio that lives or dies on one sweeping functional claim is now a fragile portfolio. These choices begin at the very start of the process; our guide on preparing an invention disclosure for your patent attorney explains how to capture the structural detail that later supports a robust claim ladder.

Invest in the specification. Because enablement is judged against the four corners of the disclosure, the specification must work harder than ever: more working examples, more structure-function correlation, more guidance that lets a POSITA reach the full claimed scope without a research program of their own. In antibody and biologic cases especially, structural claiming (sequences, CDR definitions) is far safer than purely functional claiming.

Reassess existing broad claims before asserting them. A patentee sitting on broad genus or functional claims should pressure-test them under Amgen before filing suit. An assertion built on a claim that will not survive a § 112 challenge is worse than no assertion — it invites a declaratory-judgment counterstrike and a fee award. Conversely, accused infringers should add full-scope enablement to their standard invalidity checklist; post-Amgen it is one of the most powerful defenses available against broad claims.

Watch the written-description cousin. Enablement travels with the written-description requirement, also in § 112(a), which asks whether the specification shows the inventor actually possessed the full claimed invention as of filing (Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co., 598 F.3d 1336 (Fed. Cir. 2010) (en banc)). The two doctrines reinforce each other against over-claiming, and Amgen has energized both.

The trajectory is clear and continuing: courts and the PTAB are applying Amgen aggressively to functional and genus claims, and the drafting community has responded by narrowing and structurally anchoring claims. The open questions concern exactly how much disclosure is "enough" in unpredictable arts — a fact-bound inquiry that will be litigated case by case for the foreseeable future.

Part Four: The Eligibility Stalemate — Section 101 After Alice and Mayo

What patent eligibility means, in plain terms

Before you ever reach novelty, obviousness, or enablement, a claimed invention has to clear a threshold question: is it even the kind of thing the patent system protects? That is patent eligibility under 35 U.S.C. § 101, which on its face allows patents for any "new and useful process, machine, manufacture, or composition of matter." Courts have long read three implicit exceptions into that broad language: you cannot patent a law of nature, a natural phenomenon, or an abstract idea. You cannot patent gravity, the chemical composition of a leaf, or the abstract concept of hedging risk.

The issue: the Alice/Mayo two-step and its discontents

The modern test comes from two Supreme Court cases. In Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U.S. 66 (2012), and Alice Corp. Pty. Ltd. v. CLS Bank International, 573 U.S. 208 (2014), the Court adopted a two-step framework:

  • Step one: Are the claims "directed to" one of the patent-ineligible concepts (a law of nature, natural phenomenon, or abstract idea)? If not, the claims are eligible and the inquiry ends.
  • Step two: If the claims are directed to such a concept, do they nonetheless recite "significantly more" — an "inventive concept" — that transforms the abstract idea into a patent-eligible application, rather than merely claiming the idea itself plus routine, conventional activity?

The framework's purpose, the Court explained, is to prevent patents from preempting the basic tools of scientific and technological work. Our in-depth article on patent eligibility after Alice traces this framework's application to software and business methods.

The problem is that the test is maddeningly unpredictable in application. "Directed to" and "abstract idea" have no crisp definitions, and the Federal Circuit has produced two lines of cases that are difficult to reconcile. On the pro-eligibility side, claims that recite a concrete improvement to computer functionality or to a technological process survive: Enfish, LLC v. Microsoft Corp., 822 F.3d 1327 (Fed. Cir. 2016) (self-referential database table); McRO, Inc. v. Bandai Namco Games America Inc., 837 F.3d 1299 (Fed. Cir. 2016) (specific rules for automating lip-sync animation); DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245 (Fed. Cir. 2014) (solution "necessarily rooted in computer technology"); Core Wireless Licensing S.A.R.L. v. LG Electronics, Inc., 880 F.3d 1356 (Fed. Cir. 2018) (improved display interface for small screens); Koninklijke KPN N.V. v. Gemalto M2M GmbH, 942 F.3d 1143 (Fed. Cir. 2019) (improvement to data-transmission error checking); and SRI International, Inc. v. Cisco Systems, Inc., 930 F.3d 1295 (Fed. Cir. 2019) (specific network-intrusion-detection technique). On the other side, claims that merely automate a long-standing human practice or recite a result without a specific technical means routinely fall. The line between "an improvement in technology" and "an abstract idea implemented on a computer" can turn on claim drafting as much as on substance.

The diagnostic-method context is even harsher. After Mayo, claims that detect a natural correlation — for example, that a particular biomarker level indicates a disease — have been especially vulnerable, because the underlying correlation is treated as a law of nature and the detection steps are often deemed conventional. The Federal Circuit's Vanda Pharmaceuticals Inc. v. West-Ward Pharmaceuticals, 887 F.3d 1117 (Fed. Cir. 2018), offered a partial path: a claim reciting a method of treatment — adjusting a drug dose based on a patient's genotype — was held eligible because it did more than state the natural relationship; it applied that relationship in a specific therapeutic regimen. But Vanda and Mayo are difficult to reconcile, and diagnostics without a treatment step remain a graveyard, as the en banc Federal Circuit's refusal to revisit the issue in Athena Diagnostics, Inc. v. Mayo Collaborative Services, LLC, 927 F.3d 1333 (Fed. Cir. 2019) (denial of rehearing en banc), made painfully clear — eight separate opinions, near-unanimous frustration, and an open plea for the Supreme Court or Congress to act.

The status: no Supreme Court rescue, no congressional fix

For years the patent bar has hoped the Supreme Court would take a case to clarify or recalibrate Alice/Mayo. It has repeatedly declined. The Court denied certiorari in American Axle & Manufacturing, Inc. v. Neapco Holdings LLC — widely seen as the best vehicle for revisiting eligibility — even after inviting the Solicitor General's views, who recommended granting review. Subsequent petitions (including the type of software-eligibility dispute the original version of this article flagged among the cases to watch) have met the same fate. The trajectory is one of continued judicial inaction: the Supreme Court has signaled it is content to leave Alice/Mayo in place.

Congress has tried. The proposed Patent Eligibility Restoration Act (PERA), introduced by Senators Tillis and Coons, would eliminate the judicial exceptions and replace them with a statutory list of excluded categories. As of this writing it has not been enacted. The result is a durable stalemate: the test everyone criticizes remains the governing law, and the only branches realistically positioned to change it have not acted.

The IP-strategy implication

Because the doctrine is unlikely to be rescued from outside, the burden falls on drafters and litigators to manage eligibility risk within the existing framework.

Draft for the technical-improvement narrative. The single most important lesson of the post-Alice case law is that how you frame and claim an invention often determines eligibility. Claims and specifications that articulate a specific technical problem and a specific technical solution — improving how a computer stores data, how a network detects intrusions, how an interface displays information — fare far better than claims that recite a desired result or automate a business practice. Build that improvement story into the specification from the start, with detail about what the prior art could not do and how the claimed architecture changes it.

Recite specific structure and avoid pure functional/result claiming. Just as in the enablement context, generality is the enemy. Claims that recite concrete steps, data structures, and technical mechanisms survive Step one or supply the Step two inventive concept; claims that recite "a processor configured to [achieve a result]" invite a § 101 dismissal at the pleadings.

Use the right tool for the right invention. Some software and business-method innovations are simply safer protected as trade secrets than as patents, precisely because § 101 risk is high and a trade secret never has to clear it. Algorithms, data-processing pipelines, and back-end methods that competitors cannot reverse-engineer are often better kept confidential. Our overview of the legal protection of software across copyrights, patents, trade secrets, and contracts maps these trade-offs, and our guide to building a trade-secret protection program from scratch explains how to make that election defensible. For diagnostics, pair any natural-correlation discovery with a concrete treatment or actionable application step, following Vanda's logic.

Plead and move strategically in litigation. Section 101 is frequently raised on a motion to dismiss or for judgment on the pleadings, before claim construction and before discovery — which makes it a cheap, early-exit weapon for defendants. Patent owners should anticipate this and, where possible, plead factual allegations (about unconventionality, about technical improvement) that create a fact dispute precluding early dismissal, as the Federal Circuit allowed in Aatrix Software, Inc. v. Green Shades Software, Inc., 882 F.3d 1121 (Fed. Cir. 2018), and Berkheimer v. HP Inc., 881 F.3d 1360 (Fed. Cir. 2018). Defendants should still consider raising eligibility early, while remaining alert to the Berkheimer fact-issue trap.

The trajectory here is, frankly, more of the same: continued unpredictability, continued reliance on careful drafting, and continued hope — so far unrealized — for a legislative or Supreme Court fix.

Part Five: How the Supreme Court Now Decides IP Cases — Vidal v. Elster

Anyone betting on a Supreme Court rescue from Alice/Mayo should study how the Court actually handled the one major intellectual-property case it did decide in this window. Vidal v. Elster, 602 U.S. 286 (2024), is technically a trademark case, not a patent case — but it is a window into the current Court's method, and that method has direct implications for patent owners hoping the Justices will reshape doctrine.

The issue and the holding

The case concerned the Lanham Act's "names clause," 15 U.S.C. § 1052(c), which bars registration of a trademark that consists of or comprises a living person's name without that person's consent. Steve Elster sought to register "TRUMP TOO SMALL" for shirts; the USPTO refused under the names clause. Elster argued the refusal violated the First Amendment, pointing to the Court's recent decisions striking down the Lanham Act's bars on "disparaging" marks (Matal v. Tam, 582 U.S. 218 (2017)) and "immoral or scandalous" marks (Iancu v. Brunetti, 588 U.S. 388 (2019)) — decisions we discuss in our piece on the Washington Redskins trademark troubles and the aftermath of the disparagement clause.

The Court unanimously upheld the names clause, but its reasoning is the headline. Unlike the disparagement and scandalous-marks bars, the names clause is viewpoint-neutral — it applies regardless of whether the mark praises or criticizes the named person. The controlling rationale, authored by Justice Thomas, declined to apply traditional First Amendment scrutiny to a content-based-but-viewpoint-neutral trademark-registration condition and instead grounded the result in history and tradition: there is a long, unbroken historical pedigree of restricting the trademark use of others' names, and that tradition was sufficient to sustain the clause. Several Justices concurred but objected to resting the outcome on history alone rather than a workable doctrinal test — a methodological fault line worth watching.

Why a patent strategist should care

Two lessons carry over. First, the current Court is comfortable resolving IP questions by reaching back to historical practice rather than refining modern multi-factor tests — the same instinct that, in the patent context, animated Amgen's reliance on O'Reilly v. Morse and The Incandescent Lamp Patent. A litigant asking this Court to overhaul Alice/Mayo should be prepared to argue from the historical scope of the "useful arts" and the statutory text, not merely from the doctrine's unworkability. Second, Elster confirms that this Court takes IP cases sparingly and decides them narrowly. The pattern across Amgen, the American Axle denial, and Elster is of a Court willing to police clear over-claiming and to settle discrete constitutional questions, but disinclined to author the sweeping eligibility recalibration the patent bar keeps requesting. For planning purposes, treat the Alice/Mayo status quo as durable. For the trademark-registration fallout of Elster itself, see our broader treatment of the subject matter of trademark law and the substantive standards for trademark protection.

Part Six: The Agency in Motion — The USPTO's Evolving Guidance

Patent strategy is not made only in courts. The USPTO interprets the case law for its examiners through guidance documents that, while not binding on courts, profoundly shape how applications are examined day to day. Around 2024 the agency was busy on several fronts.

Eligibility guidance, including for artificial intelligence

The USPTO maintains examination guidance implementing Alice/Mayo for examiners, including a structured framework (the agency's "2019 Revised Patent Subject Matter Eligibility Guidance" and subsequent updates) that channels the abstract-idea inquiry into enumerated groupings — mathematical concepts, certain methods of organizing human activity, and mental processes — and a practical-application analysis. The guidance is generally regarded as somewhat more applicant-friendly than the raw case law, because it gives examiners a more disciplined path to finding eligibility where claims integrate a judicial exception into a practical application.

In July 2024 the agency issued an AI-focused update to its subject-matter-eligibility guidance, complete with examples addressing how § 101 applies to artificial-intelligence and machine-learning inventions. The thrust mirrors the broader case law: claims reciting a specific improvement to the functioning of a computer or to another technology — for instance, a particular AI architecture that solves a concrete technical problem — are more likely eligible than claims that merely apply generic machine learning to automate an abstract task. This examination guidance arrived alongside the agency's February 2024 inventorship guidance for AI-assisted inventions, which confirmed that a natural person must make a "significant contribution" for any claim to be patentable, a downstream application of the Federal Circuit's holding in Thaler v. Vidal, 43 F.4th 1207 (Fed. Cir. 2022), that an AI system cannot be a named inventor. For the inventorship side of AI, see our discussion of AI and inventorship and who owns what the machine creates.

Obviousness-type double patenting and patent term

The agency and the courts were also wrestling with how the judge-made doctrine of obviousness-type double patenting (OTDP) interacts with statutory patent term adjustment (PTA). OTDP prevents an inventor from obtaining a second patent on an obvious variant of their own earlier-patented invention in a way that would effectively extend the monopoly. PTA, by contrast, adds time to a patent's term to compensate for USPTO delays. The two can collide when patents in the same family end up with different expiration dates. In In re Cellect, LLC, 81 F.4th 1216 (Fed. Cir. 2023), the Federal Circuit held that OTDP is measured against a patent's actual expiration date including any PTA — meaning a later-expiring family member can be invalidated for double patenting over an earlier-expiring relative even where the later expiration results from statutory term adjustment. The Supreme Court declined to review Cellect, leaving that rule in force and making terminal disclaimers (a tool by which an applicant disclaims the portion of a later patent's term that extends beyond an earlier related patent) a critical portfolio-management lever. The Federal Circuit later softened the edges in Allergan USA, Inc. v. MSN Laboratories Private Ltd., 111 F.4th 1358 (Fed. Cir. 2024), holding that a first-filed, first-issued patent in a family cannot be invalidated by a later-issued, later-expiring continuation — a meaningful limit on how far Cellect reaches.

Continuation and discretionary practice

The USPTO periodically floats proposals affecting continuation applications, terminal-disclaimer practice, and the Director's authority to review PTAB decisions — each of which can meaningfully change portfolio economics. The 2024 period saw proposed rulemakings touching terminal disclaimers and PTAB discretion, several of which drew heavy public comment and did not survive in their original form. The most notable was a controversial proposal to tie terminal-disclaimed patents together such that all would become unenforceable if any single claim in a linked patent were held invalid — a proposal that drew intense criticism and was ultimately withdrawn, a reminder that not every floated rule becomes law.

The IP-strategy implication

Prosecute to the guidance, litigate to the case law. Because examiners follow USPTO guidance, applicants should affirmatively map their claims and arguments to the agency's eligibility framework during prosecution — invoking the practical-application analysis and the AI examples where relevant. But never forget that a court is not bound by that guidance; a patent that squeaked through under generous agency guidance can still fall to a stricter judicial application of Alice/Mayo. Draft for both audiences. Our guide to responding to patent office actions walks through how to invoke the guidance effectively during prosecution.

Manage your family's term and OTDP exposure deliberately. Post-Cellect (as tempered by Allergan), audit patent families for OTDP risk where PTA has created staggered expiration dates, and use terminal disclaimers thoughtfully — recognizing that a terminal disclaimer can both cure OTDP and tie patents together for enforceability purposes. The interaction of OTDP, PTA, and reissue is technical and consequential.

Stay close to rulemaking. Because agency rules on continuations and PTAB discretion directly affect filing strategy, portfolio owners benefit from monitoring USPTO rulemaking and, where the stakes warrant, participating in notice-and-comment. As the withdrawn terminal-disclaimer proposal shows, comments matter.

Part Seven: The PTAB and IPR — The Fintiv Discretionary-Denial Saga

What IPR is, in plain terms

The America Invents Act of 2011 created a streamlined, administrative way to challenge an issued patent's validity without a full federal-court trial: inter partes review (IPR), conducted before the Patent Trial and Appeal Board (PTAB), a panel of administrative patent judges within the USPTO. In an IPR, a challenger (the "petitioner") argues that one or more claims are unpatentable as anticipated (§ 102) or obvious (§ 103) based on prior-art patents and printed publications. IPR is faster and cheaper than district-court litigation, uses a lower burden of proof (preponderance of the evidence, versus the clear-and-convincing standard in court), and has historically been a powerful tool for accused infringers. For how IPR coordinates with parallel court litigation, see our comprehensive guide to patent infringement litigation.

A petition does not automatically proceed. The PTAB must first decide whether to institute the review. And here is where the saga lives: the statute gives the Director (and by delegation the Board) discretion over institution — including discretion to deny a petition even when it presents a strong invalidity case.

The issue: discretionary denial under Apple v. Fintiv

In Apple Inc. v. Fintiv, Inc., IPR2020-00019 (PTAB Mar. 20, 2020) (precedential), the Board announced a set of factors for deciding whether to discretionarily deny institution because of a parallel district-court (or ITC) proceeding addressing the same patent. The Fintiv factors weigh, among other things: whether the court has granted a stay; the proximity of the court's trial date to the Board's projected final-decision deadline; the investment already made in the parallel proceeding; the overlap of issues; whether the petitioner and the court defendant are the same; and other circumstances bearing on efficiency.

The rationale was to avoid duplicative, wasteful adjudication and to respect a fast-moving district court that might resolve validity first. But critics — including many accused infringers and several major technology companies — argued Fintiv gave patent owners a way to escape IPR entirely simply by filing in a "rocket docket" with an early trial date, undermining the AIA's purpose of providing an efficient validity check. The most-criticized scenario was denial based purely on a scheduled trial date that, in practice, often slipped.

The status: a pendulum that keeps swinging

This area has been unusually volatile because it is governed largely by agency policy, which changes with leadership. The trajectory looks like a pendulum:

  • 2020–2022: expansion. Fintiv and its progeny led to a meaningful number of discretionary denials, especially for petitions tied to litigation in fast-trial districts.
  • June 2022: restriction. The USPTO Director issued interim guidance sharply curtailing Fintiv denials. The guidance provided that the Board would not discretionarily deny institution where the petition presented "compelling" evidence of unpatentability; would not rely on a trial date alone but would consider the median time to trial; and would not deny based on a parallel ITC proceeding (because the ITC cannot invalidate a patent). The guidance also clarified that a petitioner's stipulation not to pursue in court the same invalidity grounds raised in the IPR (a so-called Sotera stipulation, after Sotera Wireless, Inc. v. Masimo Corp., IPR2020-01019 (PTAB Dec. 1, 2020) (precedential)) weighs strongly against denial. This made institution substantially more available to challengers.
  • 2024–2025: re-expansion and formalization. Agency policy has since moved back toward broader discretionary denial. The USPTO rescinded the 2022 interim guidance and adopted new procedures — including a bifurcated process that routes discretionary-denial questions to the Director (or a delegated panel) for a separate, threshold determination before the merits panel ever reaches the petition. The agency also expanded the discretionary toolkit with theories such as "settled expectations" (longer-issued patents may warrant more deference) and heightened scrutiny of petitioner conduct. The net effect is a more denial-friendly posture once again, narrowing the window for challengers in cases with advanced parallel litigation.

Layered on top of Fintiv are other discretionary doctrines: denial under General Plastic Industrial Co. v. Canon Kabushiki Kaisha, IPR2016-01357 (PTAB Sept. 6, 2017) (precedential), for serial/follow-on petitions against the same patent; and denial under 35 U.S.C. § 325(d) where the same or substantially the same prior art or arguments were already considered by the Office, applying the framework of Advanced Bionics, LLC v. MED-EL Elektromedizinische Geräte GmbH, IPR2019-01469 (PTAB Feb. 13, 2020) (precedential).

The Supreme Court's decision in United States v. Arthrex, Inc., 594 U.S. 1 (2021), reinforced the Director's authority to review PTAB decisions — which is precisely why discretionary-denial policy can swing so sharply with each change in agency leadership. The institution decision itself is largely insulated from appeal under Thryv, Inc. v. Click-to-Call Technologies, LP, 590 U.S. 45 (2020), which makes the Board's (and Director's) discretion all the more consequential. Because there is effectively no judicial check on a discretionary denial, the only realistic forum for contesting one is within the agency — through Director review or rehearing — which is itself governed by the very policy that produced the denial.

A worked example

Imagine, hypothetically, a startup sued for patent infringement in a fast-trial district with a trial date eleven months out. Its instinct is to take its time, finish its prior-art investigation, and file a polished IPR petition in nine months. Under the re-expanded Fintiv regime, that delay is fatal: by the time the petition is filed, the parallel litigation will be far advanced and the court's trial will precede the Board's projected final-written-decision deadline, all but guaranteeing a discretionary denial. The strategically correct move is the opposite — file the IPR within weeks of being served, before the litigation has consumed resources, and accompany it with a Sotera stipulation foreclosing duplicative invalidity arguments in court. The same petition, same prior art, same merits — but the timing decides whether the Board ever reaches them.

The IP-strategy implication

The central planning reality is that whether IPR is even available may depend on agency policy at the moment you file — and that policy is a moving target.

For challengers (accused infringers). Speed is everything. The single most reliable way to avoid a Fintiv denial is to file the IPR petition early — ideally well before the parallel litigation has invested heavily and before trial is imminent — so the Board's projected final-decision date precedes the court's trial. Offering a Sotera stipulation remains a strong hedge against discretionary denial across policy regimes. Challengers should also avoid serial petitions that invite a General Plastic denial and should bring their best prior art that the Office has not already considered, to sidestep § 325(d). And because policy can tighten, do not assume a window that exists today will exist next year.

For patent owners. The flip side is that forum and timing on the litigation side can shape IPR exposure. Filing infringement suits in fast-trial districts has, in some policy regimes, increased the odds of a Fintiv denial of a challenger's IPR — a litigation-strategy lever that interacts directly with PTAB practice. Patent owners should also be ready to argue the discretionary factors affirmatively when opposing institution, and to pursue Director review of unfavorable institution or final-written decisions where warranted.

For everyone. Because the institution decision is largely unappealable and policy-dependent, both sides should track USPTO leadership signals and guidance the way they track case law. The trajectory is genuinely uncertain: the pendulum has swung both ways within five years, and the only safe assumption is that it will keep swinging.

Part Eight: Patent Damages — The March Toward Apportionment

Why damages drive strategy

A patent is only worth enforcing if the remedy justifies the cost. Patent damages therefore sit at the center of every enforcement decision. Under 35 U.S.C. § 284, a prevailing patentee recovers damages "adequate to compensate for the infringement, but in no event less than a reasonable royalty." Damages generally take one of two forms: lost profits (available where the patentee can prove it would have made the infringer's sales but for the infringement, under the Panduit test) or, far more commonly, a reasonable royalty — the royalty the parties would have agreed to in a hypothetical, arm's-length negotiation at the time infringement began. (This is the utility-patent regime; recall from Part Two that design-patent damages run on the separate, far more generous § 289 total-profit track.)

The issue: apportioning value to the patented feature

The reasonable-royalty analysis is built on the fifteen Georgia-Pacific factors from Georgia-Pacific Corp. v. United States Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970), which reconstruct that hypothetical negotiation. But the dominant theme of modern damages law — and the trend that continued through 2024 — is apportionment: the principle that the royalty must be tied to "only those damages attributable to the infringing features," and may not capture value contributed by unpatented components of a multi-component product (VirnetX, Inc. v. Cisco Systems, Inc., 767 F.3d 1308, 1326 (Fed. Cir. 2014)).

Apportionment matters enormously for the products that drive most modern litigation — smartphones, cars, network equipment — because a single patented feature is usually a tiny part of an enormously valuable whole. The law's tools for enforcing apportionment include:

  • The smallest salable patent-practicing unit (SSPU). Rather than basing a royalty on the revenue of an entire complex product, the royalty base should generally be the smallest salable component that practices the patent, and even then must be further apportioned if that component itself includes unpatented features (VirnetX, 767 F.3d at 1327–28; Finjan, Inc. v. Blue Coat Systems, Inc., 879 F.3d 1299, 1310–11 (Fed. Cir. 2018); Ericsson, Inc. v. D-Link Systems, Inc., 773 F.3d 1201, 1226–27 (Fed. Cir. 2014)).
  • The entire market value rule (EMVR). Only in the narrow case where the patented feature is proven to drive consumer demand for the entire product may the patentee use the whole product's revenue as the base. This is a demanding, rarely satisfied exception, not a default.
  • Daubert gatekeeping. District judges aggressively police damages-expert methodology under Daubert, excluding royalty theories not reliably tied to the case's facts — a trend traceable to Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292 (Fed. Cir. 2011), which rejected the "25 percent rule of thumb," and reinforced in Commonwealth Scientific & Industrial Research Organisation v. Cisco Systems, Inc. (CSIRO), 809 F.3d 1295 (Fed. Cir. 2015).

The status: continued tightening, with comparable licenses ascendant

The trajectory through and beyond 2024 is one of continued judicial discipline on damages. The Federal Circuit has repeatedly vacated large awards for failure to apportion or for unreliable expert methodology, and district courts have grown more willing to exclude damages experts at the Daubert stage. The court underscored the point by taking the issue en banc in EcoFactor, Inc. v. Google LLC, where the full court addressed how rigorously trial judges must scrutinize the reliability of a damages expert's royalty-rate opinion before it reaches the jury — a signal that Daubert gatekeeping in damages is tightening, not relaxing. At the same time, comparable real-world licenses have become the most persuasive evidentiary anchor for a royalty, because a sufficiently comparable license inherently reflects market apportionment — though the Federal Circuit insists that the comparison account for technological and economic differences between the benchmark license and the hypothetical one. Enhanced damages for willful infringement remain available but discretionary under Halo Electronics, Inc. v. Pulse Electronics, Inc., 579 U.S. 93 (2016), which abolished the rigid Seagate test and restored district-court discretion to punish egregious, willful infringement.

The IP-strategy implication

Build the apportionment story into the patent and the licensing record. The patentee who wins large damages is the one who can credibly tie the patented feature to a measurable slice of product value. That work starts long before litigation: documenting the technical contribution of the invention, generating comparable licenses with clean, apportioned royalty structures, and — where the facts support it — building evidence that the patented feature drives consumer demand (the EMVR predicate). A portfolio backed by a history of arm's-length comparable licenses is worth materially more in litigation than one without.

Match the asserted claims to the value. Because apportionment shrinks recoveries on multi-component products, patentees should think hard about which patents and claims to assert — favoring those whose contribution to product value is largest and most provable. Our deeper treatment of these issues appears in the comprehensive guide to patent infringement litigation.

For accused infringers, attack the base and the expert. The defense playbook is well established: challenge the royalty base (push toward the SSPU and further apportionment), attack any reliance on whole-product revenue or the EMVR, and move under Daubert to exclude unreliable royalty methodologies before they ever reach a jury.

Mind the licensing feedback loop. Every license a company grants becomes potential evidence in future litigation — for itself and against it. Structuring licenses with defensible, apportioned royalty bases protects future damages positions. See our guidance on how to license a patent from valuation to term sheet. For technologies governed by FRAND commitments, where apportionment and royalty-stacking concerns are especially acute, see our analysis of standard-essential patents and FRAND licensing.

The trajectory is settled and continuing: apportionment is entrenched, expert gatekeeping is rigorous, and the most reliable path to a defensible award runs through comparable licenses.

Part Nine: The Specific Cases on the 2024 Docket — A Watch List

The thematic parts above trace the forces reshaping patent strategy. But strategy is made one docket at a time, and when the original version of this piece was written in mid-2024, a concrete set of individual cases — some at the Supreme Court on certiorari, some pending merits decisions at the Federal Circuit, one teed up for possible en banc rehearing — embodied those forces. They are worth cataloguing both as a snapshot of the moment and because each one maps onto a theme already developed above. We treat them as a watch list, flagging the question presented and the strategic stake rather than pretending to a box score the calendar has since overtaken.

At the Supreme Court (certiorari stage)

Cellect, LLC v. Vidal (No. 23-1231) — OTDP versus statutory patent term. This certiorari petition asked the Court to review the Federal Circuit's In re Cellect holding (discussed in Part Six) that obviousness-type double patenting is measured against a patent's actual expiration date including patent term adjustment, so that a later-expiring family member can be invalidated over an earlier-expiring relative even when the later expiration flows from statutory PTA the USPTO is required to grant. The stake for portfolio owners is direct: if PTA-extended patents are vulnerable to OTDP, families built around staggered expiration dates need re-auditing and disciplined terminal-disclaimer practice. The Court declined review, leaving Cellect in force — the outcome Part Six describes.

Chestek PLLC v. Vidal (No. 23-1217) — USPTO rulemaking procedure. Though it arises from a trademark application, this case carries cross-cutting significance for patent applicants because it concerns the USPTO's rulemaking authority generally. The agency adopted a rule requiring applicants to provide a personal domicile address; a pro se applicant challenged it on privacy grounds and argued the USPTO failed to follow proper notice-and-comment procedures under the Administrative Procedure Act. The Federal Circuit upheld the rule, reasoning the domicile-address requirement was a procedural rule exempt from notice-and-comment. The strategic stake: how tightly the USPTO must adhere to APA procedure bears on every future patent rule — including the continuation, terminal-disclaimer, and PTAB-discretion rulemakings discussed in Part Six. A ruling demanding stricter procedure would slow new rules but widen the public-comment window; a permissive ruling lets the agency move faster, for better or worse.

Eolas Technologies, Inc. v. Amazon.com, Inc. (No. 23-1184) — another run at Alice/Mayo. This was one of the software-eligibility petitions that, as Part Four notes, asked the Court to revisit the much-criticized Alice/Mayo two-step. The strategic lesson is the one Part Four and Part Five draw: the Court has shown no appetite to rewrite eligibility doctrine, so drafters must manage § 101 risk within the existing framework rather than wait for rescue.

United Therapeutics Corp. v. Liquidia Technologies, Inc. (No. 23-1298) — the scope of IPR review. This petition asked whether the Federal Circuit should defer to agency discretion or instead conduct de novo review when deciding if the PTAB adjudicated a claim unpatentable on grounds or printed publications outside the scope of the IPR petition as instituted. The stake connects to Part Seven: a more deferential standard strengthens the Board's hand and makes IPR outcomes more final, while de novo review gives patent owners a stronger appellate lever against the Board — changing the risk calculus for challengers deciding whether to invest in an IPR at all.

At the Federal Circuit (pending merits)

REGENXBIO Inc. v. Sarepta Therapeutics, Inc. (No. 24-1408) — product-of-nature eligibility for engineered cells. A Delaware district court held that a cultured host cell carrying a recombinant nucleic acid molecule was not patent-eligible because the recombinant molecule's nucleic-acid sequences do not "alter" or "change" their natural form — a natural-phenomenon problem distinct from the diagnostic-method and enablement issues in Parts Three and Four, but cut from the same § 101 cloth. The patentee argues the engineered cell is markedly different from anything in nature because it combines DNA from two organisms into an artificial molecule with unique structure and function. A ruling against eligibility would narrow protection for gene-therapy and genetic-engineering inventions; a ruling for it would reinforce that human-engineered biological constructs can clear § 101. This is the Myriad/Chakrabarty "markedly different" line of the eligibility doctrine, applied to modern cell engineering.

Merck Sharp & Dohme B.V. v. Aurobindo Pharma USA (No. 23-2254) — the "issue date" of a reissued patent. The district court held that a reissued patent inherits the "issue date" of the surrendered original; Merck argues the issue date should be the date the reissued patent actually issued. The point is technical but consequential for patent-term and PTE calculations and for the timing of litigation — it ties directly into the term-management concerns of Part Six. Siding with Merck could effectively extend the usable life of certain reissued patents; affirming the district court narrows the payoff of reissue in some postures.

Alnylam Pharmaceuticals, Inc. v. Moderna, Inc. (No. 23-2357) — claim construction in chemistry. Arising out of COVID-19 mRNA-vaccine technology, this appeal turns on the construction of the claim terms "branched alkyl" and "branched C10-C20 alkyl." The district court's narrow construction produced a judgment of noninfringement for Moderna. The case is a vivid reminder — orthogonal to the validity themes above but central to enforcement — that claim construction frequently decides high-stakes pharmaceutical and biotech suits, and that precise claim drafting at prosecution is the only reliable hedge against an adverse Markman ruling years later.

Regents of the University of California v. Broad Institute, Inc. (No. 22-1594) — CRISPR-Cas9 inventorship. The long-running dispute over who invented the use of CRISPR-Cas9 in eukaryotic cells continued, with UC challenging the PTAB's determination that the Broad Institute was first to invent in that setting. UC argues the Board applied the wrong conception standard and that UC reduced the system to practice first. Beyond the extraordinary value of the patent rights at issue, the case bears on how conception and inventorship are proven for complex biotech inventions — and it pairs naturally with the AI-inventorship guidance discussed in Part Six and our companion pieces on AI and inventorship and who owns what the machine creates. The practical lesson is evergreen: contemporaneous documentation of the inventive process is what wins inventorship fights.

Arlton v. AeroVironment, Inc. (No. 21-2049) — government-contractor immunity under § 1498. The plaintiffs alleged that AeroVironment's Mars Helicopter (the NASA Ingenuity rotorcraft) infringed their "Rotary Wing Vehicle" patent. The district court granted summary judgment under 28 U.S.C. § 1498, which channels patent claims arising from work "for the United States" into the Court of Federal Claims and immunizes the contractor from an ordinary infringement suit. The appeal contests the scope of that immunity. A broad reading shields government contractors and encourages government work; a narrow reading exposes them to infringement suits — a meaningful consideration for any company in the defense, aerospace, or government-technology supply chain.

Micron Technology, Inc. v. Longhorn IP LLC (No. 23-2007) — state anti-patent-trolling laws and federal preemption. An Idaho district court, invoking an Idaho statute aimed at abusive patent assertion, required the patent assertor to post an $8 million bond to pursue infringement claims against Micron. The appellant argues the Idaho Act is unconstitutional and preempted by federal patent law; Idaho and some two dozen other states defend it. The outcome could either embolden a patchwork of state anti-troll statutes or curtail state regulation of patent assertion and renew pressure for a federal solution — a structural question about who gets to regulate patent-enforcement conduct.

Pending possible en banc rehearing

Apple Inc. v. Omni Medsci, Inc. (No. 23-1034) — the limits of PTAB discretion over reply arguments. A Federal Circuit panel held that the PTAB erred in disregarding an Apple argument about a claim limitation in the prior art that Apple raised in response to Omni's patent-owner response. Omni sought rehearing en banc, contending the panel's decision improperly cabined the Board's discretion to refuse new reply arguments. This is a procedural cousin of the Fintiv and § 325(d) discretion themes in Part Seven: how much latitude the Board has to police the scope of an IPR shapes how dynamic — and how predictable — these proceedings are for both sides.

The strategic through-line of the watch list

These eleven matters are not a random assortment; they cluster around the same pressure points the thematic parts identify. Two are about eligibility (Eolas, REGENXBIO) and the persistent § 101 uncertainty of Part Four. Three are about PTAB and IPR mechanics (United Therapeutics, Apple v. Omni Medsci, and the discretionary backdrop of Part Seven). Two are about patent term and prosecution technicalities (Cellect, Merck v. Aurobindo) that feed the term-management discipline of Part Six. The rest probe the institutional edges of the system — agency rulemaking (Chestek), claim construction (Alnylam), inventorship (UC v. Broad), sovereign-work immunity (Arlton), and the federal-state boundary (Micron). The durable takeaway is not how any single one came out, but that an IP portfolio has to be built to withstand challenges arriving from all of these directions at once.

Part Ten: Synthesizing the Trajectory — A Strategic Playbook

Step back from the individual doctrines and a coherent picture emerges. The patent system around 2024–2025 was moving, across multiple fronts at once, toward demanding more precision from patentees and offering more avenues to challengers. Design patents lost their special protection from obviousness while keeping their devastating § 289 remedy. Broad functional claims lost ground to enablement. Software and diagnostic eligibility stayed treacherous, and the Supreme Court — as Elster confirmed — is in no mood to rewrite the rules. Damages stayed apportioned and tightly policed. The one area that swung toward patent owners — discretionary IPR denials — did so unstably, dependent on agency politics. The throughline is that vague, broad, and lazily drafted patents are worth less and survive less often than they used to, while carefully scoped, well-documented, well-supported patents retain — and even increase — their value.

That diagnosis yields a durable playbook for portfolio owners:

Draft narrowly and deeply, then ladder upward. Whether the doctrine is enablement, eligibility, or obviousness, the recurring lesson is that specificity wins. Claim concrete structures, specific technical mechanisms, and provable improvements; support them with rich specifications and working examples; and build claim ladders so that if a broad claim falls, narrower claims survive. Generality is now a liability across the board.

Document the non-obvious and the valuable contemporaneously. LKQ makes objective indicia central to defending design patents; apportionment makes value-attribution evidence central to damages; eligibility makes the technical-improvement narrative central to surviving § 101. In every case, the evidence is far more persuasive when generated in real time — press coverage, awards, comparable licenses, technical white papers — than when reconstructed during litigation.

Diversify your protection. Where § 101 risk is high or claims would have to be drawn broadly to be useful, consider trade-secret protection as a complement or alternative to patenting. The right answer is often a layered strategy: utility patents for what can be reverse-engineered and reliably claimed, design patents for ornamental features (with their potent § 289 remedy), trade secrets for the rest, and contracts to backstop everything. Our guide to protecting a mobile app with a comprehensive IP strategy shows that layered approach applied to a single product.

Plan litigation and PTAB strategy together. IPR availability, forum selection, trial timing, and discretionary-denial policy are now interlocking variables. Accused infringers should file IPRs early and offer Sotera stipulations; patent owners should weigh fast-trial forums and Director review; and both should monitor agency policy as closely as they monitor case law.

Audit existing portfolios against the new law. Many patents granted under older standards — broad genus claims pre-Amgen, design patents secured under Rosen-Durling, software claims drafted without a technical-improvement story — are weaker than their owners assume. A periodic portfolio audit against current doctrine identifies which assets to assert, which to license, which to abandon at the next maintenance-fee deadline, and which to shore up through continuation practice while the family is still alive. For the freedom-to-operate side of the same analysis, see conducting a freedom-to-operate analysis for new products.

Frequently Asked Questions

Did the Supreme Court ever fix the Alice/Mayo eligibility mess? No. As of this writing the Court has repeatedly declined to revisit § 101 eligibility, most prominently by denying certiorari in American Axle. Legislative proposals such as the Patent Eligibility Restoration Act have not been enacted. And Vidal v. Elster signaled a Court that decides IP questions narrowly and by reference to history and tradition, not one inclined to author a sweeping eligibility overhaul. The practical consequence is that the unpredictable two-step test remains the law, and careful claim drafting — emphasizing a specific technical improvement — is the principal way to manage eligibility risk.

Are design patents now worthless after LKQ? Not at all. LKQ made design patents easier to challenge for obviousness by replacing the rigid Rosen-Durling test with the flexible Graham framework, but design patents remain valuable — they are still relatively cheap, still support the unique total-profit remedy of § 289 (which after Samsung v. Apple turns on identifying the relevant "article of manufacture"), and can still be defended with strong objective indicia of non-obviousness (industry praise, commercial success, copying). The change calls for more deliberate prosecution and evidence-building, not abandonment of design protection.

What is the Samsung v. Apple "article of manufacture" rule, and why does it matter? Under 35 U.S.C. § 289 a design-patent infringer owes the patentee its total profit on the relevant "article of manufacture." In Samsung Electronics Co. v. Apple Inc., 580 U.S. 53 (2016), the Supreme Court held that the article need not be the entire product sold to consumers — it can be a component. The Court left lower courts to work out how to identify that article in a multicomponent product, so the size of a § 289 award often turns on whether the relevant article is the whole product or a small part of it. That single question can move a damages award by orders of magnitude.

Does Amgen v. Sanofi mean I can no longer get broad patent claims? No. Amgen did not ban broad, functional, or genus claims; it reaffirmed that a claim must enable its full scope without undue experimentation. You can still claim broadly if your specification genuinely teaches a POSITA how to make and use the whole claimed scope. The safe practice is to pair any broad claim with progressively narrower claims anchored to specific structures and working examples, so that the portfolio survives even if the broadest claim is invalidated.

What is the single best way for an accused infringer to avoid a Fintiv discretionary denial of an IPR? File the IPR petition early — well before the parallel district-court litigation has advanced and before trial is imminent — and offer a Sotera stipulation agreeing not to pursue in court the same invalidity grounds raised in the IPR. Both steps cut strongly against discretionary denial across the agency's shifting policy regimes. Be aware, though, that the agency's 2024–2025 procedures route discretionary-denial questions to the Director or a delegated panel as a threshold matter, so the timing pressure on petitioners is greater than ever.

Why are patent damages awards getting harder to maximize? Because of apportionment. Courts require that a reasonable royalty reflect only the value of the patented feature, not the value of the whole multi-component product. Tools like the smallest-salable-patent-practicing-unit rule, the narrow entire-market-value-rule exception, and rigorous Daubert scrutiny of damages experts — reinforced by the en banc EcoFactor v. Google proceedings — all push awards toward the value actually attributable to the invention. Comparable real-world licenses have become the most reliable way to anchor a defensible royalty. Note that this apportionment regime governs utility patents; design-patent damages run on the separate § 289 total-profit track.

How often should I review my patent portfolio against these developments? At minimum, review the portfolio whenever a maintenance fee comes due (so you do not pay to keep weak patents alive) and before asserting any patent in litigation. A broader strategic audit — assessing enablement, eligibility, obviousness, and OTDP/term exposure across the portfolio against current law — every year or two is prudent for any company whose business depends on its patents.

Key Takeaways

  • LKQ v. GM overruled the Rosen-Durling test and now judges design-patent obviousness under the same flexible Graham/KSR framework used for utility patents; certiorari was denied and the USPTO updated examiner guidance accordingly, making design patents easier to challenge and elevating the importance of objective indicia.
  • Design patents remain potent because § 289 lets the owner recover the infringer's total profit on the relevant article of manufacture; after Samsung v. Apple, the size of that recovery turns on the unresolved question of how to identify that article in a multicomponent product.
  • Amgen v. Sanofi reaffirmed that broad functional and genus claims must enable their full scope without undue experimentation, driving a shift toward narrower, structurally anchored claims supported by richer specifications.
  • Section 101 eligibility remains governed by the unpredictable Alice/Mayo two-step, with no Supreme Court or congressional fix in sight; Vidal v. Elster confirmed a Court that decides IP cases narrowly and historically, so the technical-improvement narrative, careful drafting, and trade secrets are the realistic defenses.
  • The USPTO's evolving guidance — including its 2024 AI eligibility and AI-inventorship guidance and the post-Cellect/Allergan terrain on double patenting and patent term — shapes prosecution day to day, even though courts are not bound by it.
  • The Fintiv discretionary-denial saga makes IPR availability a moving, policy-dependent target; with the agency's 2024–2025 swing back toward denial, challengers should file early and stipulate, while patent owners can use forum and timing as levers.
  • Patent damages continue their march toward strict apportionment and rigorous expert gatekeeping (underscored by the en banc EcoFactor proceedings), making value-attribution evidence and comparable licenses central to any utility-patent enforcement strategy.
  • The unifying lesson is that precision, documentation, and diversification of IP protection now determine portfolio value far more than breadth alone.

Related Articles

This article is provided by mclaw.io for general informational purposes only. It is not legal advice, does not create an attorney-client relationship, and may not reflect the most current legal developments, which continue to evolve. Patent law is fact-specific and jurisdiction-specific; readers should consult qualified patent counsel before making decisions about their own intellectual property.