Patent Docs

Agilent Technologies, Inc. v. Synthego Corp. (Fed. Cir. 2025)

By Kevin E. Noonan --

An argument could be made that one of the most significant Supreme Court decisions in U.S. patent law in the last thirty years was Dickinson v. Zurko.  In that case the Court held that the Federal Circuit was bound by the provisions (5 U.S.C. § 706) of the Administrative Procedures Act to defer to factual determinations made by administrative agencies within the area of their expertise.  (It should be understood that the Supreme Court's more recent decision in Loper Bright Enterprises v. Raimondo did not disturb this statutory standard, being limited to agency interpretations of the laws under which they were established or administered.)  The effects of this interpretation of the laws governing judicial review can be observed (and have been amplified) in the Federal Circuit's opinions regarding Patent Trial and Appeal Board decisions, particularly as applied in inter partes review (IPR).  The most recent application of this standard can be found in Agilent Technologies, Inc. v. Synthego Corp.

Synthego challenged two Agilent patents in its IPRs:  U.S. Patent Nos. 10,337,001 and 10,900,034. These patents were directed to CRISPR-Cas9* systems for gene editing.  Claim 1 in each of these patents are set forth in the opinion:

The '001 patent:

A synthetic CRISPR guide RNA having at least one 5′-end and at least one 3′-end, the synthetic guide RNA comprising:
    (a) one or more modified nucleotides within five nucleotides from said 5′-end, or
    (b) one or more modified nucleotides within five nucleotides from said 3′-end, or
    (c) both (a) and (b);
    wherein said guide RNA comprises one or more RNA molecules, and has gRNA functionality comprising associating with a Cas protein and targeting the gRNA:Cas protein complex to a target polynucleotide, wherein the modified nucleotide has a modification to a phosphodiester linkage, a sugar, or both.

(emphasis added).

The '034 patent:

A synthetic CRISPR guide RNA comprising:
    (a) a crRNA segment comprising (i) a guide sequence capable of hybridizing to a target sequence in a polynucleotide, (ii) a stem sequence; and
    (b) a tracrRNA segment comprising a nucleotide sequence that is partially or completely complementary to the stem sequence,
    wherein the synthetic guide RNA has gRNA functionality comprising associating with a Cas protein and targeting the gRNA:Cas protein complex to the target sequence, and comprises one or more modifications in the guide sequence, wherein the one or more modifications comprises a 2′-O-methyl.

(emphasis added).

The opinion characterizes as "[t]he key prior art" supporting Synthego's challenge on anticipation grounds to be an International Application, published as International Publication No. WO 2015/026885, filed on August 20, 2014 by Pioneer Hi-Bred International. This application disclosed "compositions and methods" for performing CRISPR featuring "guide polynucleotide[s]["gRNA"]" in both single- and dual-molecule forms; specific disclosure included Example 4, for modifying the nucleotide base, phosphodiester bond linkage or molecular topography of the guiding nucleic acid component(s) of the guide polynucleotide/Cas endonuclease system" "for increasing cleavage activity and specificity."  Also disclosed were modified RNA molecules comprising "nucleotide and/or phosphodiester bond modifications [that] may be introduced to reduce unwanted degradation."  With regard to Synthego's contentions of invalidity for obviousness, two additional references were asserted:  a first scientific publication by Threlfall et al. (2011, Synthesis and Biological Activity of Phosphonoacetate- and Thiophosphonoacetate-modified 2′-O-methyl Oligoribonucleotides, 10 Org. Biomol. Chem. 746–54) and a second scientific publication by Deleavey et al. (2012, Designing Chemically Modified Oligonucleotides for Targeted Gene Silencing, 19 Chem. & Bio. Review 937–54).  The Threlfall reference discloses specific nucleotide modifications using PACE (phosphonoacetate)- or thioPACE (phosphonothioacetate)-modified phosphate groups, and the Deleavey reference discloses "a vast array" of nucleotide modifications to stabilize oligonucleotides (including RNA oligonucleotides).

The PTAB found all claims in each patent to be unpatentable.**  The Board found that the Pioneer Hi-Bred reference disclosed a functional gRNA and was enabling, and that those claims that were not anticipated were obvious.  This appeal followed.

The Federal Circuit affirmed, in a decision by Judge Prost, joined by Judges Linn and Reyna.  In announcing the "substantial evidence" standard for the Board's factual determinations (which included anticipation, citing St. Jude Med., LLC v. Snyders Heart Valve LLC, 977 F.3d 1232, 1238 (Fed. Cir. 2020)), the opinion relies on In re Bayer Aktiengesellschaft for the principle that "[w]here two different conclusions may be warranted based on the evidence of record, the Board's decision to favor one conclusion over the other is the type of decision that must be sustained by this court as supported by substantial evidence," 488 F.3d 960, 970 (Fed. Cir. 2007) (emphasis added).

Agilent propounded three arguments in its appeal:  first, that the Board's anticipation determination was not supported by substantial evidence; second, that the Pioneer Hi-Bred application was not enabling; and third, that the skilled worker would not have had a reasonable expectation of success (with regard to the Board's obviousness determinations) that the PACE and thioPACE modifications would have been successfully used to modify gRNA in a CRISPR-Cas9 system.  Regarding Agilent's first argument, the panel determined that the Board's anticipation decision was supported by substantial evidence.  The opinion enumerates the facts the Board relied upon concerning the Pioneer Hi-Bred disclosure:

"Pioneer Hi-Bred discloses that the guide polynucleotides described therein can: (1) form a complex with a Cas endonuclease; and (2) enable the endonuclease to recognize a DNA target site.  That disclosure reads on both the associating and targeting aspects of the 'gRNA functionality' recited" in the challenged claims.

Examples 4 and 5 of the Pioneer Hi-Bred disclosure were specifically called out in the opinion for referring to "modified guide nucleotides," which the Board interpreted as "indicating that those sequences have [the claimed gRNA] functionality," supported by several additional statements in the Pioneer Hi-Bred disclosure.  The Board had rejected (and the Federal Circuit affirmed as being supported by substantial evidence) Agilent's argument that the purported lack of cleavage activity using such modified gRNA nucleotide "demonstrate[d] that Pioneer Hi-Bred does not disclose gRNA functionality," based in part on admissions by Agilent's expert that "just because a gRNA in Table 4 [in Pioneer Hi-Bred] lacks cleavage activity does not demonstrate that it also lacks the ability to bind a Cas protein and target that complex to target polynucleotide."  Taken in its totality, the panel held that the Board's decision that the Pioneer Hi-Bred application anticipated certain of the claims of the '001 and '034 patents was supported by substantial evidence.

The opinion then turns to Agilent's argument that the disclosure of the Pioneer Hi-Bred application was not enabling, where the panel "[saw] no error" in the Board's conclusion that the reference was enabling.  One basis for the Court's decision in this regard was recognition that for assertion as an anticipating reference only an enabling disclosure, not evidence of actual reduction to practice is required, under inter alia, Schering Corp. v. Geneva Pharms., 339 F.3d 1373, 1380 (Fed. Cir. 2003); Rasmusson v. SmithKline Beecham Corp., 413 F.3d 1318,1326 (Fed. Cir. 2005); and In re Morsa, 803 F.3d 1374, 1376 (Fed. Cir. 2015) (cited in the opinion).  Moreover "[p]rior art disclosures are presumed enabling" under Impax Labs., Inc. v. Aventis Pharms., Inc., 545 F.3d 1312, 1316 (Fed. Cir. 2008), according to the opinion.  The panel appreciated that the Board performed its enablement assessment under the conventional rubrics set forth in In re Wands, 858 F.2d 731 (Fed. Cir. 1988), and that the Pioneer Hi-Bred disclosure passed muster under those standards (while recognizing that certain disclosures, including Examples 4 and 5 were prophetic).  Regarding the relative novelty of CRISPR-Cas9 systems when the Pioneer Hi-Bred application was filed the Board stated that:

[W]hile the art was somewhat unpredictable in December 2014, it was far from a blank slate with a [person of ordinary skill in the art] understanding how the different elements of a CRISPR/Cas system are used and function together, including the role of gRNA; the types of chemical modifications that had been successfully used in other systems to reduce RNA degradation, while preserving functionality; and standard techniques for making gRNAs with the modifications disclosed and exemplified in Pioneer Hi-Bred.

The Federal Circuit in affirming the Board's holding that the Pioneer Hi-Bred disclosure was enabling rejected Agilent's argument that the factual circumstances in the Court's Impax decision were analogous.  There the issue was enablement of a disclosure of "hundreds or thousands of compounds and several diseases," as well as "broad and general" dosage guidelines "without sufficient direction or guidance to prescribe a treatment regimen, whereas here "Pioneer Hi-Bred exemplifies particular crRNA sequences having the recited chemical modifications at the recited locations and teaches that gRNA comprising such may be used as guide polynucleotides in a CRISPR Cas system."  In addition the Board had supported its enablement holding because "the particular types of chemical modifications disclosed in Pioneer Hi-Bred and recited in the challenged claims had been known and used for decades to stabilize RNA against unwanted degradation in other systems."  Also unavailing was Agilent's reliance on the Supreme Court's Amgen Inc. v. Sanofi decision, the opinion noting distinctions "in two meaningful ways."  First, the Court's Amgen decision involved "whether the asserted claims were sufficiently enabling to be valid under 35 U.S.C. § 112, not whether a prior-art reference was enabling and could thus support anticipation," which are "two separate inquiries," citing Novo Nordisk Pharms., Inc. v. Bio-Tech. Gen. Corp., 424 F.3d 1347, 1355 (Fed. Cir. 2005).  According to the panel, enablement under Section 112 requires a disclosure to enable the skilled worker to be able to use the claimed invention, whereas for Section 102 purposes no such disclosure of utility is required.  Second, the Board found that the Pioneer Hi-Bred disclosure would have been understood by the person of ordinary skill in the art with regard to "how the different elements of a CRISPR/Cas system are used and function together, including the role of gRNA; the types of chemical modifications that had been successfully used in other systems to reduce RNA degradation, while preserving functionality; and standard techniques for making gRNAs with the modifications disclosed and exemplified in Pioneer Hi-Bred" and that this disclosure was enabling.  In addition, the Federal Circuit rejected Agilent's argument regarding Pioneer Hi-Bred's purported disclosure of "many inoperable guide [RNAs]" because the data Agilent relied upon for this argument related to synthetic DNA sequences not modified RNA recited in the challenged claims.  The last specific Agilent argument rejected by the Board and affirmed by the panel was that the Pioneer Hi-Bred disclosure did not enable a "single guide RNA" (sgRNA), which according to the opinion was expressly disclosed in the reference.

Finally, the opinion assessed and rejected Agilent's arguments that the Board erred in finding certain other claims in each patent to be invalid for obviousness.  The grounds for these arguments was that the Pioneer Hi-Bred disclosure "did not expressly disclose the functionality of the claimed PACE- or thioPACE-modified guides" and that the Board did not explain its findings regarding there being a reasonable expectation of success.  Regarding the former argument, the opinion states that Agilent conflated the requirements for anticipation with those for obviousness, wherein the lack of PACE- or thioPACE-modified guides disclosure in the Pioneer Hi-Bred reference was remedied by combination with the Threlfall and Deleavey disclosures, as supported by expert testimony adduced by Synthego.  Regarding Agilent's latter argument, the panel considered as "thorough" the Board's analysis that the person of ordinary skill would have reasonably expected success from the combination of the Pioneer Hi-Bred reference with either the Threlfall or Deleavey disclosures.  The opinion sets forth the Board's reasoning to illustrate its decision to affirm:

[B]y December 2014, several studies had shown that the CRISPR/Cas system could successfully tolerate modifications.  While these studies describe different types of modifications than those in the challenged claims, such evidence nevertheless supports Dr. Furneaux's testimony that a [person of ordinary skill in the art] would have expected that chemical modifications could be made at the 5[′] and 3[′]-ends of a gRNA while preserving the Cas enzyme's gene editing function.

The record further demonstrates that shortly after the discovery of the CRISPR/Cas system for gene editing and prior to December 2014, there were already a number of researchers in addition to the authors of the Pioneer Hi-Bred publication suggesting the use of the claimed chemical modifications to improve the resistance of gRNA to degradation.  [Agilent's] expert, Dr. Marshall, conceded as much on cross-examination.  The fact that multiple groups of researchers independently suggested the same types of gRNA modifications re cited in the challenged claims evidences that a [person of ordinary skill in the art] would have had a reasonable expectation those modifications could be successfully employed in a CRISPR/Cas system.  Moreover, while [Synthego] points to multiple references suggesting such modifications to gRNA, neither [Agilent] nor Dr. Marshall identify any reference expressing doubt that such modifications could be successfully implemented in a CRISPR/Cas system.  This contrast undermines [Agilent's] argument that a [person of ordinary skill in the art] would not have reasonably expected the prior art modifications to work in a CRISPR/Cas system.

This record was sufficient for the Federal Circuit to affirm the Board's decision as being supported by substantial evidence.

Agilent Technologies, Inc. v. Synthego Corp. (Fed. Cir. 2025)
Panel: Circuit Judges Prost, Linn, and Reyna
Opinion by Circuit Judge Prost

* The opinion in a footnote informs that "CRISPR" stands for "clusters of regularly interspaced short palindromic repeats."

** Claims 1–7, 9, 10, 12–15, 17, 18, 20–25, and 27–30 of the '001 patent and claims 1–5, 8–21, and 24–33 of the '034 patent were anticipated, and claims 8, 11, 16, 19, and 26 of the '001 patent and claims 6, 7, 22, and 23 of the '034 patent were obvious over the cited references.

Subject Matter Eligibility in the 21st Century: Echoes of pre-§ 103 Obviousness*

By Kevin E. Noonan --

The evolution of subject matter eligibility after the Supreme Court's decisions in Prometheus v. Mayo, Alice v. CLS Bank, and Association for Molecular Pathology v. Myriad Genetics has resulted in a regime of predictable unpredictability and certain uncertainty (a recent cogent treatment of the phenomenon in computer science technology patenting can be found in "The Narrow Pathway to Patent Eligibility in the Federal Circuit").  The situation is reminiscent of the state of obviousness in mid-Twentieth century prior to enactment of § 103 of the patent statute, when whether a patented invention was protected was a matter of judicial subjectivity.  As a consequence and as recognized by Justice Jackson, "the only patent which is valid is one which this court has not been able to get its hands on," Jungerson v. Ostby, 335 U.S. 560, 571 (1949).

A brief survey of seminal obviousness cases illustrated this similarity:

• Hotchkiss v. Greenwood (1850) standard:  an inventor's invention must have "more" than what an "ordinary or skillful mechanic" would provide.

• Cuno Engineering Corp. v. Automatic Devices Corp. (1941):  the inventive act had to come into the mind of an inventor in a "flash of genius" and not as a result of tinkering.  "The new device, however useful it may be, must reveal the flash of creative genius, not merely the skill of the calling."

• Graham v. John Deere (1966):  interprets the statute to conform to the Court's requirements that a patentable invention must be distinguished over the relevant prior art.

• Sakraida v. Ag Pro, Inc. (1976):  necessary to show some "synergy" between those known elements and denied patentability because the invention "simply arrange[d] old elements with each performing the same function it had been known to perform."

• Sakraida expressly disregarded by Federal Circuit in Stratoflex, Inc. v. Aeroquip Corp. (Fed. Cir. 1983):  "[a] requirement for 'synergism' or a 'synergistic effect' is nowhere found in the statute . . . virtually all patents are 'combination patents,' if by that label one intends to describe patents having claims to inventions formed of a combination of elements."

• "Teaching, Suggestion, or Motivation" (TSM) test developed by Federal Circuit, set aside by Supreme Court in KSR v. Teleflex but based on Graham and tied to the statute and prior art.

Even the semantics of subject matter eligibility (SME) determinations and their analysis conjures the subjectivity and uncertainties of obviousness; how more defined or reliable is the existence of an "inventive concept" and the "flash of genius," "synergy," or something "more" that the endeavors of an "ordinary mechanic"?  The history of how the Patent Office, Patent Trial and Appeal Board, numerous district courts, and the Federal Circuit have implemented the two-part Mayo/Alice test shows that subjectivity and unreliability.  It isn't far-fetched to consider that these consequences are not definitive for the same reasons obviousness without reliance on prior art was subject to subjective considerations (hindsight in obviousness); it is often more productive to recognize a "know it when you see it" approach ala Justice Potter Stewart (on entirely different circumstances and subject matter).  Indeed, one of the organizing principles (to the extent they can be understood) of subject matter ineligibility is often based on overbroad claiming (as in Morse v. O'Reilly, a case cited in support of SME deficiencies that had been better recognized for over a century as having failed to satisfy the disclosure requirements under Section 112).

The history of Federal Circuit jurisprudence (that has led some members of that Court to bemoan the outcome of these cases; see Judge Moore's dissent in Athena Diagnostics, Inc. v. Mayo Collaborative Services, LLC ) are set forth below:

There are some certainties in these cases, nevertheless.  For example, diagnostic methods are almost per se ineligible.  There has been some eligibility traction in formulating claims based on natural laws in the context of a novel assay employing the natural law in certain situations; however, the consequences can be relying on new assays for each circumstance and avoiding "routine, well-understood, and conventional" assays.  More frequently claims are drafted in the context of a method of treatment informed by the results of diagnostic assay (see "Natural Alternatives International, Inc. v. Creative Compounds, LLC (Fed. Cir. 2019)" and "Endo Pharmaceuticals Inc. v. Teva Pharmaceuticals USA, Inc. (Fed. Cir. 2019)").  Such claims are susceptible to two well-recognized caveats:  inherent prior art anticipation (when individuals identified as treatment recipients received drug prior to development of the diagnostic assay; novel pharmaceuticals can avoid this outcome), and the "divided infringement" problem, when infringement is effectuated by separate practitioners of the recited limitations of the treatment method (wherein, for example, who administers the test is different from who administered the treatment and either actor cannot be characterized as being under the "direction and control" of the other).  The most coherent calculus for method claims to pass SME muster was enunciated by Judge Lourie in his Vanda decision, illustrating the advantages of claiming specifically and thereby avoiding overbreadth:

At bottom, the claims here are directed to a specific method of treatment for specific patients using a specific compound at specific doses to achieve a specific outcome.  They are different from Mayo.  They recite more than the natural relationship between CYP2D6 metabolizer genotype and the risk of QTc prolongation.  Instead, they recite a method of treating patients based on this relationship that makes iloperidone safer by lowering the risk of QTc prolongation.  Accordingly, the claims are patent eligible [emphasis added].

Similarly, natural products are per se generally ineligible, although Cellzdirect and Illumina decisions suggest isolated natural product having specific utility not possessed by product in nature may be eligible when properly/adequately claimed.  Generally, formulations and combinations not arising in nature can be eligible and while databases of biological information are often per se ineligible, their use may be (as in XY).

An Illustration of when formulations have failed to be eligible is Chromadex v. Elysium Health (Fed. Cir. 2023) (which is explained at least in part as being granted in 2005, prior to the imposition of current standards of SME).  The claims of U.S. Patent No. 8,197,807 are directed to dietary supplements comprising nicotinamide ribonucleotide, converted in vivo to NAD+, a co-enzyme involved in redox reactions:

Claimed were compositions formulated with other components as set forth in the table (reproduced in the opinion), which had in common many similarities to milk except for the limitation that the nicotinamide ribonucleotide ([NR]) was isolated.  The Federal Circuit's analysis resembled in many respects an obviousness assessment and characterized the claims as "reading on" milk with the exception that [NR] is "isolated."  But under Myriad being "isolated" is not enough (in contrast to the state of SME when these claims were prosecuted under, inter alia, In re Bergy).  The opinion also noted that the claims were "very broad," consistent with overbreadth being a hallmark of claims failing to recite eligible subject matter, and with the distinctions with Natural Alternatives Int'l, Inc. v. Creative Compounds, LLC, where the Court recognized that claims to a natural product (beta-alanine) were patent-eligible for reciting "specific treatment formulations that incorporate[d] natural products" and that those formulations "ha[d] different characteristics and c[ould] be used in a manner that beta-alanine as it appears in nature cannot."

To the extent a claiming strategy can be elucidated from these cases it might be to assiduously avoid overbroad claiming, based on current application of Section 112 criteria (Judge Lourie's Vanda specificity example).  A useful analogy might be to adopt a "pointillist approach" that avoids overbreadth and preemption by reciting a plurality of narrow claims directed to specific embodiments, as illustrated by Georges Seurat's "A Sunday Afternoon on the Island of La Grande Jatte" (also analogous to Twentieth Century television transmission; see "The Last Lone Inventor" for further descriptions).

There is another way to look at the issue of SME and the 'judicial exceptions" contained in Shahshahani, "Against the abstract ideas exclusion," 40 Berkeley Technology Law Journal (forthcoming 2025).  In this paper the rationales for SME are applied to conventional abstract ideas and natural laws and found wanting.  Rather than accepting the conventional wisdom for the need to protect inventions that are "building blocks" due to perceived risks of inhibiting future innovation (by preemption), Professor Shahshahani argues that such building block inventions are important to incentivize by patents because their recognition and the application of natural laws and abstract ideas provides the capacity for development of future inventions employing them.  One illustration of this analysis is that while natural law often predates human discovery of it and could fall under the eligibility limitations in Parker v. Flook, human knowledge of the law provides the opportunity and ability to apply such natural laws and thus the applications thereof should not be precluded from eligibility.  The deficiencies of the conventional rationale for the judicial exceptions are further developed by recognizing that evidence supporting the conventional wisdom regarding preemption is not always correct, citing evidence that patents on genes did not inhibit genetic research, as shown by Bhaven Sampat & Heidi Williams, How Do Patents Affect Follow-On Innovation? Evidence from the Human Genome, 109 AMER. ECON. REV. 203, 217-19, 232 (2019) (existence of "gene patents" did not inhibit genetic research ) and Janet Freilich & Sepehr Shahshahani, Measuring Follow-On Innovation, 52 RES. POL'Y 104854 (2023) (finding that gene patents close to expiration caused an increase in follow-on research but those that were far from expiration had no effect).  This scholarship suggests that a reevaluation of the premises upon which the exceptions (natural laws, natural phenomena, abstract ideas) are based ("tying up" and thereby inhibiting fundamental building blocks necessary for future innovation) should be reconsidered for not being valid and not having the deleterious effects ascribed to them.  While perhaps not being (at the present time) a strategy for successfully arguing patent eligibility before the PTO or a district court in the absence of the Supreme Court addressing SME concerns (having denied certiorari to do so in almost 100 cases), further empirical evidence on the continued viability of SME restrictions may someday spur a necessary and beneficial reassessment.

* Taken from a presentation at the American Conference Institute's 23rd Annual Advanced Summit on Life Sciences Patents, presented on May 19th, 2025 in New York City.

Dancing with Abstract Ideas: Patent Eligibility in 2025

By Michael Borella --

In June 2014, the U.S. Supreme Court handed down Alice Corp. v. CLS Bank Int'l, establishing a now-infamous two-step, judicially-imposed test for patent subject-matter eligibility that narrowed the broad statutory eligibility principles set forth in 35 U.S.C. § 101.  Under Alice, one must first determine whether a patent claim is directed to an ineligible concept such as an abstract idea, a law of nature, or a natural phenomenon.  If so, the second step requires further determination of whether the claim as an ordered combination amounts to significantly more than just the ineligible concept.  In theory, this framework was meant to prevent the sweeping preemption of fields of technology that ostensibly should remain in the public domain.  In practice, however, the Alice test has proven exceedingly difficult to apply consistently, especially for software inventions.  Despite software having a concrete manifestation on physical media and being used to change the state of tangible devices, it often gets lumped into the abstract idea category even when claimed so narrowly that it can only preempt a specific use case.

The disarray is such that even older precedents that were once shining examples of eligibility have been thrown into doubt.  The Supreme Court's own 1981 Diamond v. Diehr decision, which upheld the eligibility of a patent on a computer-implemented rubber curing process, was reaffirmed as good law in Alice.  Yet, the Alice test as currently applied could easily be used to argue that Diehr's claims are actually ineligible.[1]  This uncertainty has real-world consequences, raising the risk and cost for small and mid-sized companies seeking to protect their innovations.

If Alice set the stage, the U.S. Court of Appeals for the Federal Circuit -- the court that hears all patent appeals from district courts and the United States Patent and Trademark Office (USPTO) -- has written the chapters that followed.  The last decade has seen the Federal Circuit dramatically narrow the scope of patent-eligible software, decision by decision.  First, many business methods were declared ineligible, then more and more software inventions implemented on a general-purpose computer, then even software on any platform, and eventually some non-software innovations got swept up as well.

This tightening chokehold is reflected in the Federal Circuit's own track record.  In 2024, for example, the Federal Circuit decided 22 patent cases on substantive grounds of § 101 and found the claims eligible in only one.[2]  In other words, a staggering 95.5% effective invalidity rate of patents on appeal.  For patent owners, the Federal Circuit has become an abattoir for software patents, with only the rarest of claims surviving.

On the front lines of patent eligibility, the USPTO has tried to bring some stability to the chaos.  The USPTO periodically issues guidance to help examiners (and applicants) navigate the morass of § 101 case law.  A notable effort was the January 2019 Revised Patent Subject Matter Eligibility Guidance, which categorized abstract ideas into three buckets (mathematical concepts, methods of organizing human activity, and mental processes) and instructed examiners to consider a claim to recite an abstract idea only if it falls into one of those recognized categories.  The 2019 guidance also introduced the notion of considering whether a claim involving an abstract idea integrates it into a practical application.  If the claimed invention is applied in a specific manner that improves technology or another field, it might be deemed patent-eligible at step one despite involving an abstract idea.  This gave applicants new arguments to overcome rejections by pointing to technological benefits in their inventions.

By many accounts, the 2019 guidance somewhat reined in the wild inconsistency of examiner decisions and led to a modest increase in software patent allowances . . . for a while.[3]  However, unpredictability still rules the USPTO's application of Alice, especially across different examiner art units.  Some examiners (particularly in business-method oriented units) continue to issue near-automatic § 101 rejections, while others in computer hardware, software, and communications art units might be more receptive if the applicant articulates a clear technical improvement.  Appeals to the Patent Trial and Appeal Board (PTAB) are often an uphill battle -- recent statistics show the PTAB affirms examiners' § 101 rejections about 90% of the time.[4]  In effect, unless you can fit your claim into the mold of a past case found eligible or an example from USPTO guidance, the safer bet is that the PTAB will say "no patent for you."

The subjective nature of Alice leaves room for reasonable minds to differ.  And they do.  The experience of practicing before the USPTO can feel like trying to shoot at a moving target.  Even after the 2019 guidance, applicants faced divergent outcomes on similar inventions.  Examiners have significant discretion in characterizing what a claim is "directed to" at step one and what counts as "conventional" at step two.  Thus, a savvy applicant must preemptively stack the deck in their favor through careful drafting (more on that below).

In late 2023 and 2024, the USPTO began grappling with how emerging technologies like artificial intelligence (AI) fit into the eligibility framework.  Under a mandate from former President Biden's 2023 Executive Order on AI, the USPTO issued an Updated Subject Matter Eligibility Guidance in July 2024 focusing on AI-related inventions.  Many had hoped this update -- the first in almost five years -- would address some of the widely voiced concerns about the unpredictability of examining AI and software inventions. Unfortunately, the 2024 Guidance dashed those hopes.[5]

The update largely restated generic eligibility principles and added only a few AI-specific examples.  Even worse, the guidance arguably doubles down on a hardline approach suggesting that advances in AI and uses of AI are likely ineligible unless they are implemented in specific hardware circuitry or cause a significant change to the state of an ancillary system.  In other words, merely improving an AI algorithm in software might not satisfy the USPTO's idea of a practical application unless you tie it to a particular machine or a real-world effect, even if that improvement saves massive amounts of compute and storage capacity on the algorithm's host computer.

The Supreme Court, for its part, has repeatedly declined to hear new patent eligibility cases that might clarify the doctrine.  This includes denying certiorari in 2022's American Axle case, which many hoped would address the Federal Circuit's overly aggressive application of the law of nature exception and maybe soften Alice's impact.  In Congress, a bipartisan group of legislators has been pushing for reform.  Senators Thom Tillis and Chris Coons, among others, recently re-introduced the Patent Eligibility Restoration Act (PERA) to rewrite § 101 and roll back some of the Alice-era exclusions.  But this is not the first time that a version of PERA has been introduced, and the patent community is waiting for the bill to get much closer to being a law before giving it their full attention.

Thus, 2025 can be characterized as largely status quo in terms of patent eligibility.  Stakeholders are still watching the Supreme Court and Capitol Hill for rays of hope, perhaps a sense that someone, somewhere is going to pull the eligibility doctrine back to solid constitutional ground.  But there is no reason to believe that meaningful change will take place this year.

Given this challenging landscape, what can attorneys, applicants, and patentees do to improve their odds of surviving a § 101 challenge?  While no strategy is foolproof (indeed, even the most meticulous drafting cannot guarantee eligibility under an uncertain test), there are techniques that can tip the scales at least a modest amount.  Patent practitioners over the last decade have essentially reverse-engineered the patterns of decisions to develop a playbook for software patents.  Below are some best practices as a mix of drafting tips and prosecution strategies to help prevent or overcome a § 101 challenge.

1.  Focus on a specific technical improvement: Frame the invention as a solution to a technical problem, not a business.  Put differently, the claimed invention should improve the operation or performance of machines rather than that of businesses or people.  In the specification, clearly articulate the technological improvement or advantage the software provides.  For example, does it make a computer run faster, use memory more efficiently, use less network capacity, or operate under reduced power requirements?  Does it improve features fundamental and specific to computer technology, such as network security, data compression, parallel processing, or overall availability?  The Federal Circuit has emphasized that software claims directed to an improvement in computer functionality are not abstract.  Make sure the claims explicitly recite the features that form this improvement, and that the specification describes how those features solve a specific technical challenge.  In short, tell a story of innovation in real-world technology rather than casting the invention of an automation of a business process or human task.  In fact, try to avoid emphasizing or even mentioning any business process or human task at all.

2.  Avoid outcome-driven claim language: Draft concrete, step-by-step claims that explain how the invention achieves its results.  Purely functional or result-oriented claims (e.g., "determining a classification for the data using machine learning") are red flags.  Instead, include the specific mechanisms or steps that carry out the improvement.  If your invention uses an algorithm, spell out key steps of that algorithm rather than claiming just the end result.  If it processes data, indicate how the data is processed in a new way.  Broad and generic claims are especially vulnerable under § 101, so consider adding technical limitations.  For instance, instead of "generating a network map," you might claim a specific step like "iteratively constructing a network map based on real-time telemetry data received from network nodes."  This helps tie the idea to concrete procedures or components.

3.  Recite hardware or real-world effects (when possible): One way to show a claim is not abstract is to root it in a physical context or device.  If the software interacts with new or specialized hardware, include those details.  Even if it's running on a generic computer, mention any sensors, controllers, or other devices that are part of the system and used by the invention. Similarly, if the software produces a tangible real-world result (e.g., controlling an industrial process, adjusting a camera's settings, or providing a specific type of medical treatment), emphasize that.  Simply saying "by a computer" or "on the internet" is unlikely to help. But describing, for example, how a specific AI algorithm embedded in a medical imaging device yields clearer images for diagnosis will make the invention more concrete.

4.  Highlight any non-conventional techniques: Under Alice step two (the inventive concept analysis), claims that merely use routine or well-known computer functions won't cut it.  If your invention uses an algorithm or data processing technique that is not conventional, or combines steps in a unique way, call that out in the specification.  The goal is to create a factual basis that, if someone later alleges that the claimed invention involves just standard computing, you can rebut that by pointing to your disclosure.  After the Federal Circuit's 2018 Berkheimer decision and ensuing USPTO memos, examiners are supposed to provide evidence when they claim something is well-understood, routine, or conventional.  While this policy is not always followed strictly, it arms you with an argument:  if you have described why your solution is innovative at a technical level, you can insist that an examiner (or court) acknowledge those concrete details rather than ignoring them.

5.  Include a statement of technical improvement in the specification: In the unpredictable world of software patents, the specification is your lifeline.  Use it to explicitly teach why the invention is technologically significant.  Don't assume the benefit is obvious.  Include an explanation of the technical problem (e.g., "existing media distribution systems could not handle real-time network changes without significant adaptation delay, which led to failures in live streaming or broadcasting") and how your invention overcomes it ("the claimed embodiments provide a solution by incorporating an iterative machine learning model that dynamically updates a representation of the network in real-time, something not achievable with prior manual or static methods").  The statement of technical benefit can take various forms, but it should include descriptions of at least:  (i) the technical deficiencies in previous systems, (ii) the how the claimed invention overcomes these technical deficiencies, (iii) as many of the claimed invention's technical advantages as you can come up with (e.g., "by reducing the number of database accesses by an order of magnitude, these embodiments use less computational, network, and power capacity than conventional techniques.").  Finally, leave the door open for other technical improvements potentially flowing from the claimed invention as well as the possibility that this invention could be used to solve other technical problems not explicitly discussed in the specification.

6.  Reinforce the technical improvement at every opportunity in the specification: Even a general statement of technical improvement might not be enough.  As you describe and specify the invention, consider enhancing each and every feature with a sentence or two describing its technical improvement.  The goal is to repeatedly hit the reader over the head with technical improvements, as well as to give you a number of locations in the specification to point to during prosecution or litigation that support the patent eligibility of the claims.  One example might be, "The system pre-allocates memory blocks in a deterministic schedule based on static task profiling, avoiding dynamic memory access during execution.  This eliminates runtime memory fragmentation and latency spikes in real-time systems, improving timing predictability and reliability and thus providing a concrete enhancement in computer memory management."  Another example might be, "The scheduler dynamically assigns processor cores and their clock rates based on predicted energy consumption models for application tasks derived from historical sensor and usage data.  Doing so reduces power consumption without user-perceivable performance degradation, directly improving hardware resource management in mobile computing."

7.  Include performance results or data demonstrating an improvement: Benchmarking data, latency reductions, improved throughput, or lower resource usage can help substantiate that the invention provides a technical benefit.  However, applicants must exercise caution -- if your claims recite an abstract idea, examiners and courts may construe performance gains as improvements to abstract idea itself rather than to the underlying computing technology.  To mitigate this risk, the specification should clearly tie the performance benefits to specific technical features -- such as arrangements of software modules, hardware interactions, or architectural changes, rather than to general functional outcomes.

8.  Add narrower embodiments that include more hardware or specific steps: In case the broad version of the claims hits an eligibility roadblock, these more focused embodiments might survive.  For example, reciting specific hardware components such as a sensor, display controller, or network interface can help anchor the claim in concrete technology rather than abstract functionality.  Similarly, adding detailed steps, like preprocessing data in a defined format, performing a particular type of signal transformation, or using a designated memory allocation strategy, can demonstrate that the invention is not merely a result-oriented abstraction.  These narrower embodiments provide fallback positions during prosecution or litigation and can help show that the claimed invention improves the operation of a particular machine or system.  In some cases, it may be beneficial to include at least one very narrow claim directed to the precise implementation over which protection is sought.  This forces the examiner to engage with a concrete technical contribution of the invention rather than dismissing all possible implementations as highly general abstract ideas.  Even if such a claim has limited commercial value on its own, it can serve as a foothold for establishing eligibility and opening dialogue during prosecution.

9.  Be aware of USPTO examples and precedents: It can be helpful to draft claims that are comparable to ones that have been deemed eligible in the past.  The USPTO's examples (from 2019 and the 2024 update) provide model claims that the USPTO purports to be eligible.  If the invention is technologically similar to one of these examples, consider using language from the example in your specification.  Nonetheless, some examiners are skeptical of this USPTO guidance, as it does not have the authority of Federal Circuit case law.

10.  Don't over-rely on Enfish and McRo: While these Federal Circuit cases are still good law and provide useful guidance for arguing that software-based claims can be eligible when they improve the functioning of a computer itself, they represent relatively rare wins in a legal landscape that has grown more skeptical of software patents.  Subsequent decisions have narrowed their reach, often distinguishing them on factual grounds or finding that later claims failed to recite comparable specificity.  Courts now expect a more rigorous showing that the invention solves a technological problem in a non-generic way, preferably with supporting disclosures in the specification that demonstrate how the claimed advance is achieved and why it matters from a technical perspective.  When drafting claims and a specification, do not assume that any similarities to the claims of Enfish or McRo will be given much weight by an examiner.

11.  Always interview the examiner when you are dealing with a § 101 rejection during prosecution: Examiner interpretation of subject matter eligibility varies widely, and what one examiner considers a clearly abstract idea, another might view as a technological improvement.  These differences are often shaped by the examiner's background, personal approach to the Alice framework, and the prevailing norms of their art unit.  Different examiners and art units can have dramatically different § 101 rejection rates even within the same technical center.  An interview gives you the opportunity to understand how the examiner is construing the claims, what part of the specification they find compelling (if any), and whether they are open to particular claim amendments or additional details that could overcome the rejection.

It bears repeating that even following all these best practices is not a guarantee.  The legal standard for patent eligibility remains, to put it mildly, a work in progress.  By heeding these lessons, however, particularly the careful tailoring of claims and the emphasis on detailed technical improvements in the specification, practitioners can significantly improve their chances of securing and defending software patents in this hostile environment.

[1] https://www.patentdocs.org/2021/04/could-alice-be-used-to-invalidate-diehr-of-course-it-could.html.

[2] https://www.patentdocs.org/2025/05/the-narrow-pathway-to-patent-eligibility-in-the-federal-circuit.html.

[3] Numerous attorneys report anecdotally that the rate of § 101 rejections in 2025 has increased over that of previous years.

[4] https://www.patentdocs.org/2024/08/91-that-is-the-rate-at-which-the-ptab-affirms-examiner-section-101-rejections.html.

[5] https://www.patentdocs.org/2024/07/uspto-publishes-updated-subject-matter-eligibility-guidance-focusing-on-ai.html.

Treprostinil Saga Continues

By Kevin E. Noonan --

United Therapeutics Corp. (UTC) and Liquidia Technologies Inc. have been battling for close to a decade over their respective inhaled formulations of treprostinil (UTC's Tyvaso®, Liquidia's Yutrepia™) for treatment of pulmonary hypertension and their dispute is not yet over, both parties filing suits asserting patents against one another.

Treprostinil

In their extensive litigation, UTC has asserted claims in U.S. Patent No. 9,593,066 as well as U.S. Patent No. 10,716,793.  The District Court held that the asserted claims of the '066 patent were invalid for anticipation or were not infringed, a decision affirmed in all respects by the Federal Circuit, and that the asserted claims in the '793 patent were infringed and were not invalid (see "United Therapeutics Corp. v. Liquidia Technologies, Inc. (Fed. Cir. 2023)".  In an inter partes review proceeding, the Patent Trial and Appeal Board held that the challenged claims of the '793 patent (which the District Court held were infringed) were invalid for obviousness, a decision also affirmed by the Federal Circuit (see "United Therapeutics Corp. v. Liquidia Technologies, Inc. (Fed. Cir. 2024)).

In the more recent cases, UTC asserted (in the Middle District of North Carolina, 1:25CV368, before Judge Thomas Schroeder, complaint filed on May 9, 2025) claims of U.S. Patent No. 11,357,782 (issued June 14, 2022, which expires May 14, 2027) for methods of treating pulmonary hypertension using dry powder formulations; claim 1 is representative:

1.  A method of treating pulmonary hypertension comprising:
    providing an inhalation device for treating pulmonary hypertension in a human suffering from pulmonary hypertension comprising a powder formulation of treprostinil or a pharmaceutically acceptable salt thereof and a dry powder inhaler configured to administer single event dose of the powder formulation comprising treprostinil or a pharmaceutically acceptable salt thereof,
    wherein the single event dose comprises at least 15 micrograms to 90 micrograms of treprostinil or a pharmaceutically acceptable salt thereof delivered in 1 to 3 breaths,
    wherein the dry powder inhaler is configured to administer the entire single event dose in less than 5 minutes with at least 5 micrograms of treprostinil or a pharmaceutically acceptable salt thereof being inhaled per breath through coordinated actuation of the dry powder inhaler with each breath, and
    administering to a human suffering from pulmonary hypertension with the dry powder inhaler the single event dose comprising at least 15 micrograms to 90 micrograms of treprostinil or a pharmaceutically acceptable salt thereof in 1 to 3 breaths,
    wherein the human administers the entire single event dose with the dry powder inhaler in less than 5 minutes by inhaling at least 5 micrograms of treprostinil or a pharmaceutically acceptable salt thereof per breath by coordinating one actuation of the dry powder inhaler for each separate breath, and
    wherein administration of an additional single event dose in the same manner occurs at least 3 hours later.

An important distinction is that UTC's product, encompassed by the limitations of the '782 patent is a new formulation of UTC's Tyvaso® product called Tyvaso DPI®, which obtained FDA approval in 2022 and (according to UTC's complaint) "is the first marketed dry powder formulation of treprostinil in the United States."  Liquidia's accused product, Yutrepiaä, is also a dry powder formulation of treprostinil approved by FDA on August 19, 2024.  However, this approval came with the caveat that Liquidia's product would not get final approval until UTC's 3-year regulatory exclusivity expired on May 23, 2025, which approval was granted when that deadline recently passed (although Liquidia has not yet put its Yutrepiaä product on the market).

UTC alleges that practicing Yutrepia's label will literally infringe claim 1 of the '782 patent, as well as claims 2, 3, 8.  UTC's complaint, filed May 9, 2025 asks for a declaratory judgment of inducing infringement under 35 U.S.C. § 271(b) and contributory infringement under 35 U.S.C. § 271(c) based on Liquidia's current failure to launch their product.  UTC's allegations are based on "Liquidia's unauthorized development, manufacturing, commercial marketing, distribution, offers for sale, sales and/or use of Yutrepia™ (treprostinil) inhalation powder for the treatment of pulmonary hypertension ("PH")," based on "tentative approval" of Liquidia's Yutrepiaä product.  UTC's allegations are illustrated by this cartoon included in the complaint:

Liquidia asserts non-infringement based on UTC's claims requiring coordination between actuation of the dry powder in the inhaler and inhalation by the patient, which is not required by the Yutrepia's device/method.  Liquidia further asserts invalidity based on the PTAB's invalidation of the '793 patent based on timing (the Final Written Decision of the IPR in the '793 patent was issued "only weeks after" allowance of the '782 patent), and that a related UTC application (US 2022-0323459 A1) was rejected for obviousness based on the reasoning in the FWD in the '793 IPR.

In a separate litigation, UTC filed suit under the Hatch-Waxman Act in the District of Delaware against Liquidia's earlier Yutrepiaä product when Liquidia amended its NDA (NDA No. 213005, filed under § 505(b)(2)) for this drug to include a new indication, pulmonary hypertension associated with interstitial lung disease ("PH-ILD").  Initially UTC asserted the '793 patent (later invalidated by the PTAB in inter partes review; see above) and U.S. Patent No. 11,826,327, claim 1 being representative:

1.  A method of improving exercise capacity in a patient having pulmonary hypertension associated with interstitial lung disease [i.e., PH-ILD], comprising administering by inhalation to the patient having pulmonary hypertension associated with interstitial lung disease an effective amount of at least 15 micrograms up to a maximum tolerated dose of treprostinil or a pharmaceutically acceptable salt thereof in a single administration event that comprises at least 6 micrograms per breath.

Relevant to proceedings in the Middle District of North Carolina case (see below), UTC's motion for preliminary injunction to prevent Liquidia from marketing its Yutrepiaä product was denied in that case.  This case is set for trial this month.

In another dispute in inter partes review, the PTAB invalidated UTC's challenged claims in U.S. Patent No. 9,604,901, with claim 1 being a representative claim:

A pharmaceutical batch consisting of treprostinil or a salt thereof and impurities resulting from (a) alkylating a benzindene triol, (b) hydrolyzing the product of step (a) to form a solution comprising treprostinil, (c) contacting the solution comprising treprostinil from step (b) with a base to form a salt of treprostinil, (d) isolating the salt of treprostinil, and (e) optionally reacting the salt of treprostinil with an acid to form treprostinil, and wherein the pharmaceutical batch contains at least 2.9 g of treprostinil or its salt.

This decision was affirmed by the Federal Circuit (see "United Therapeutics Corp. v. Liquidia Technologies Inc. (Fed. Cir. 2024)").

For its part Liquidia sued UTC for infringement of its U.S. Patent No. 10,898,494, which patent issued January 26, 2021 and expires May 5, 2037 and recites representative claim 1:

1.  A method for treating a patient, comprising:
    administration of a dry powder composition comprising
    from about 100 micrograms to about 300 micrograms treprostinil or a pharmaceutically acceptable salt thereof to a patient
    by inhalation using a dry powder inhaler over one to four breaths to treat pulmonary hypertension.

Liquidia's complaint mirrors UTC's complaint but is limited to inducement and/or contributory infringement (the patient being the literal infringer by using the inhaler as prescribed).  Liquidia interprets UTC's treatment regimen containing on its label to direct patients to administer 112 micrograms (mcg) or 128 mcg at each treatment session for 4 sessions per day, which falls within the scope of UTC's asserted claims.

The most recent activity in these litigations was the District Court's denial in its North Carolina action of UTC's motions for a temporary restraining order and preliminary injunction.  In denying these motions, the District Court held that UTC failed to satisfy the requirement for a showing of a likelihood of success on the merits and that Liquidia's Yutrepiaä product market entry would cause irreparable harm.  The basis for UTC's failure to satisfy the "likelihood of success" requirement focused on whether the asserted claims were invalid for obviousness.  As explained in the opinion, in the preliminary injunction context the burden on the accused infringer is to raise the invalidity issue "that [the patentee] cannot show 'lacks substantial merit,'" citing BlephEx, LLC v. Myco Indus., Inc., 24 F.4th 1391, 1399 (Fed. Cir. 2022).  But, the District Court cautioned, "[t]he burden of presenting a substantial question of invalidity at this stage is lower than what is required to prove invalidity at trial," under Entegris, Inc. v. Pall Corp., 490 F.3d 1340, 1351 (Fed. Cir. 2007), and other authority.  Here, Liquidia's basis for asserting invalidity was the PTAB's IPR decision that the '793 patent was invalid for obviousness (a decision that issued after the Examiner allowed the application leading to the '782 patent).  Liquidia's position is that the disparate decisions create a contradiction, insofar as the '793 invalidation was based on the same prior art as the '782 allowance (a representative claim from each patent shown side-by-side below):

The Examiner for the application that resulted in the '782 patent reasoned that "a person of ordinary skill in the art would be unable to 'predict the dose, the number of breaths, and the timing of taking the drug' based on the relevant prior art," a conclusion not reached by the PTAB Board in invalidating the '793 patent (specifically the dose range, 15-90 micrograms, and delivery in "one to three" breaths).  Liquidia's invalidity argument was supported, they contended, by rejection of claims in UTC's U.S. Patent Application No. 17/745,333 for obviousness made after the PTAB's decision invalidating the '793 patent became available.  UTC's attempts to distinguish the limitations recited in these claims and the disclosure in the specification were unavailing in attempting to convince the District Court that Liquidia's obviousness arguments were unsupported by the evidence; the District Court supported its decision regarding such arguments by UTC's failure to convince the Examiner during prosecution of the '333 application using similar arguments, the comparison of the three sets of claims shown in this table:

(It being clear that claim 1 of the '782 patent is recited with significantly more specificity (i.e., limitations) than the claims of either the '793 patent or the '333 application; nevertheless these distinctions were not persuasive before the District Court.)  And in the denial of UTC's preliminary injunction motion, Judge Schroeder rejected arguments based on any differences in effect regarding the '793 and '782 claims being method of treatment claims and the '333 application claims being directed to pharmaceutical compositions, saying that "[i]f it is not the case that what constitutes an obvious inference from the prior art with respect to a pharmaceutical product is obvious as to a method of treatment, UTC has not persuasively made that case."

Turning to the "irreparable harm" prong of the requirements for a preliminary amendment, the District Court also found a lack of persuasiveness from UTC.  Rather, the Court was persuaded in part by UTC's delay in bringing suit on the '782 patent (from June 2022 until May 9, 2025, for purportedly strategic reasons, as well as a six-month delay from the date when FDA established Liquidia's Yutrepiaä product label -- the Court characterized this situation as involving "last-minute filing").  Also relevant was Liquidia's assertion that it was sufficiently "financially viable" to compensate UTC for any monetary losses it might incur after a Yutrepiaä product launch.  The Court further expressed doubts about the risk of price erosion argued by UTC, and UTC's cause was not aided by statements by UTC executives minimizing the risk of a Yutrepiaä product launch, in some instances to be less than 5% of their current revenues.

In view of UTC's failures in establishing the first two prongs of the requirements for being granted a preliminary injunction, the District Court's relatively perfunctory analysis of the other two prongs (balance of the hardships and the public interest) did nothing to change the outcome, and the Court denied the injunction of these grounds.

As a consequence of the District Court's decision, within days Liquidia announced that it was preparing for a full commercial launch of its Yutrepiaä product.  And UTC's third quarter 2024 projected annual revenues of $3 billion provided important incentives for further litigation over this market.

Microprocessor v. Abacus: The Fictional Patent-Eligibility Debate between a Patent Attorney and a Retired Supreme Court Justice

By Michael Borella --

Background:  The oral argument of Bob v. Algorithm Savings & Loan (2014), p. 6:

JUSTICE HÄAGEN-DAZS:  Imagine King Tut lounging outside his pyramid, surrounded by gold and bad financial instincts.  He's handing out chits left and right, "Good for one unit of gold, redeemable later."  He's got an abacus guy furiously moving beads to keep up.  Eventually, the abacus guy says, "Uh, dude, you've promised more gold than you actually have."  So Tut yells, "Stop!"  Crisis averted.  Or take my mother—she didn't need a computer.  Just one glance at my checkbook and she'd declare, "Hey, you've overdrawn. Stop!"  So now I ask you what, exactly, is less abstract about a computer saying "stop" instead of my mom or a Bronze Age accountant with a bead toy?

Scene:  A quiet conference room at a law and technology symposium.  An experienced patent attorney with a degree in computer engineering sits across from retired U.S. Supreme Court Justice Häagen-Dazs.  They are discussing technology in the context of patent law.

Justice Häagen-Dazs:  I've always been skeptical about software patents.  I recall, in Bob, we noted that using a computer to perform an abstract idea isn't enough.  After all, isn't a computer just a very fast abacus?

Patent Attorney:  With respect, Justice Häagen-Dazs, I would argue that's a profound oversimplification.  An abacus is a passive mechanical tool.  It requires a human operator to interpret, manipulate, and extract meaning.  It doesn't process anything on its own.  A microprocessor, on the other hand, is an active, autonomous computing engine capable of executing millions of logic operations per second according to programmed instructions.  It doesn't just assist a person's calculation, it performs the calculation.

Justice Häagen-Dazs:  But aren't both merely performing arithmetic?

Patent Attorney:  The distinction lies in functional capacity and architecture.  A microprocessor implements a general-purpose model of computation, like a Turing machine.  It can simulate any algorithmic process, given enough time and memory.  In practice, an abacus has no such versatility.  It can't run an operating system, control a spacecraft, or manage a distributed database on its own or even with the assistance of a human.  It doesn't have conditional logic, memory registers, or instruction sets.  When patent law considers eligibility, this matters because we're not just talking about number crunching.  Computers are dynamic, programmable systems that change state and manage complex workflows.

Justice Häagen-Dazs:  You mean my King Tut hypothetical from Bob was wrong?

Patent Attorney:  Your analogies conflate what is being done with how it is being done.  Yes, the goal of preventing overspending is the same.  But the mechanism for achieving that goal in a computer is entirely different from a human using common sense or manually tallying numbers.  As a more tangible example, if the only known method of binding two pieces of wood together is a nail, then that should not prevent the inventor of the screw from obtaining a patent.

Justice Häagen-Dazs:  But couldn't a human operating an abacus do anything a microprocessor could do?  Isn't it just a generic tool like a pen and paper or a handheld calculator?

Patent Attorney:  No one has ever used an abacus to stream a video or encrypt HTTPS traffic, much less train a large language model -- for good reason.  The analogy breaks down because an abacus can only be used to perform simple mathematical operations to solve small scale problems.  It would be absurd to contend that you could replace a modern graphical processor with even thousands of humans armed with abacuses.  The problems being solved and the solutions are fundamentally different.

Justice Häagen-Dazs:  But the concern I had -- and still have -- is that merely implementing a longstanding human practice on a computer shouldn't suffice for patent eligibility.  Otherwise, we'd be monopolizing abstract ideas with trivial technological dressings.

Patent Attorney:  The key question is whether the computer implementation is trivial.  If it merely automates a human practice using generic computing steps, such as storing, retrieving, and displaying, then forget about § 101 -- it's not patentable as obvious.  But in many cases, the implementation is non-trivial in that the inventors aren't just using the computer as a faster pencil.  They are solving technical problems that arise only in the context of computer implementation.  Take for example a method of organizing human activity, say, a known manual accounting method.  If someone's claim is so broad that it covers that method, it's clearly unpatentable.  But if the same concept is implemented in a way that changes how the problem is solved because of the unique constraints or capabilities of the computing environment, and further the claim reflects this difference, then the computer is inseparable from this solution.  The notion of the abstract idea from Bob fails to reflect this fundamental fact.  Going back to a pre-Bob regime of patentability would not result in thousands of low-quality patents, especially if the USPTO properly applies the principles of obviousness, functional claiming, and definiteness from the patent statute.

Justice Häagen-Dazs:  So in your view, the difference is not merely one of speed or convenience but of computational capability?

Patent Attorney:  Precisely.  A microprocessor embodies a flexible, programmable platform capable of executing a vast range of machine-readable instructions.  It's not just an electronic abacus.  It is a fundamentally different class of tool.  Comparing it to an abacus is like comparing a steam engine to a horse.

Justice Häagen-Dazs:  I see.  Perhaps I underestimated how much abstraction and logic can be embedded in hardware.  But I've been told over and over that there's something special happening under the hood in software patents, yet I can't see what's happening.  I'd rather err on the side of caution than risk giving twenty-year monopolies to ideas that should remain in the public domain.

Patent Attorney:  Um, there's this thing called "discovery" where parties can produce evidence, such as source code, system architectures, and technical declarations from programmers, to show exactly what's happening "under the hood."  The courts don't have to guess.

Justice Häagen-Dazs:  Yes, yes, I'm aware of discovery.  But even then, I see lines of code, flowcharts, declarations by experts with PhDs in computer science . . . and I still don't see a machine.  I see words.  Diagrams.  Assertions that something is "faster" or "more efficient."  To me, it's all just squiggly nonsense in a box labeled "processing unit."  It might as well be alchemy.

Patent Attorney (getting visibly annoyed):  Justice Häagen-Dazs . . . Steve . . . a microprocessor is neither alchemy nor it is an abacus.  As my Gen Z kids like to tell me, this is what happens when you let old people make the important decisions . . . .

Justice Häagen-Dazs (looking at a sundial strapped to his wrist):  Oh dear, is that the time?  I'm late for pickleball.  Toodles!

Justice Häagen-Dazs stands up and hurries toward the exit, only to collide with an attendee entering the room.  The attendee's mobile phone (which happens to be running an investment app) slips from their hand and lands squarely on Justice Häagen-Dazs' foot.  He winces and curses the "abstract idea" that just fractured his toe and cost him six weeks of pickleball.

End Scene

Federal Trade Commission Continues Efforts to Delist Drug Device Administration Patents in Orange Book

By Kevin E. Noonan --

"Plus ça change, plus c'est la même chose," Jean-Baptiste Alphonse Karr (1849)

In the last years of the Biden Administration, the Federal Trade Commission issued a policy statement and sent letters to ten companies having Orange Book-listed patents claiming devices for administering drugs challenging the propriety of those listings and demanding that the companies delist these patents (see "FTC Announces Efforts to Police Pharmaceutical Companies' Patent Behavior" and "FTC Warns Pharma Companies It Means Business with Its Orange Book Listing Policy").  The changes during the past 100+ days of the Trump Presidency could hardly be more dramatic:  Chairwoman Lina Khan and two Commissioners appointed by Democratic Presidents are gone (voluntarily or by disputed Presidential fiat) and almost any Biden initiative has been challenged or reversed; indeed, with apologies to Justice Jackson it can fairly be said that "the only [Biden initiative] which is valid is one which this [President] has not been able to get [his] hands on."

The FTC's crusade against Orange Book-listed device patents is the exception to this trend (proving that the perception that "drugs cost too much" and that patents are to blame is bipartisan, as is the seeming misunderstandings regarding the purported linkage between them).  The latest FTC action has been another spate of warning letters sent to Novartis, Mylan, Teva, Amphastar, and Covis (indicating that at least some of the previously threatened companies had earlier complied).

One difference in this latest foray is the existence of the Federal Circuit affirmance of a district court order to delist, in Teva Pharms. Inc. v. Amneal Pharms. LLC, following the doctrinal grounds relied upon by the FTC in its former incarnation.  This should come as no surprise, because the FTC filed amicus briefs before both courts, and the district court opinion was unabashed in incorporating the arguments in the FTC's brief in support of its ruling and the legal reasoning supporting it.

While the changes in philosophy between the two Administrations (and the composition of the FTC could not be more partisan), in practice the rhetoric in the letters are almost identical.  Each letter asserts that the recipient has an "ongoing obligation to ensure the propriety of its patent listings in the FDA's [Orange Book]," citing the earlier letters, all the letters expressly referenced the Federal Circuit decision in Teva v. Amneal, and in Teva's case not limiting the delisting demand to the patents at issue in that case.  The earlier public pronouncements are relegated to a footnote in these letters but were cited as a reminder nonetheless.  And all these letters were signed by Kelse Moen, Deputy Director, Bureau of Competition; Mr. Moen was appointed by FTC Chair Andrew Ferguson on February 18, 2025.

The FTC has been enunciating these threats for almost two years, and while being conventionally active in pursuing judicial remedies as in Teva v. Amneal (and previously in the series of cases resulting in FTC v. Actavis) has not followed through on its demands.

The patents challenged by the FTC for each pharmaceutical company are as follows:

The Weird Philosophies Adjacent to the AI Movement

By Michael Borella --

Love it or hate it, artificial intelligence (AI) has ushered the world to an inflection point.  The power of large language models (LLMs) is changing the structure of labor markets, education, informational gatekeeping, and governance.  It is still unclear whether these changes will lead to a utopian scenario in which all human needs are taken care of by machines or a neo-feudal hellscape in which a few ultra-wealthy technocrats cyber-rule over billions of platform-dependent serfs.  Or maybe we land somewhere in between.  Regardless of where it is ultimately headed, AI is the current big thing and on track to continue expanding.

Given this, it is important (and even mildly entertaining until you get to the end of this article) to consider the weirdness of some AI-adjacent subcultures.  While the vast majority of individuals working in the AI space do not adhere to these philosophies and beliefs, a handful of notable and highly-influential technologists are either true believers or have been known to rely on these fringe philosophies.

The rationalists are a loosely affiliated intellectual community focused on human reasoning, understanding cognitive biases, and anticipating long-term risks from advanced technologies such as AI.[1]  Many believe that artificial general intelligence (AGI) poses an existential risk unless it is aligned with human values.  This sounds fine, but some rationalists have taken this to an extreme degree, overlapping intellectually with longtermers, another esoteric niche that places the needs of future humans at or above the level of those of current humans.[2]

This has led to some strange positions, often based on self-centered, post-hoc justifications.  Longtermers place an equal moral weight on the well-being of future humans and that of current humans.  Thus, they posit that there will be many more human beings in the future than there are now, and so a utilitarian and "rational" conclusion is to optimize for the needs of those of the future even if this causes some degree of suffering in those of the present.  Sometimes these lines of thinking are referred to as "effective altruism."

Probably the most famous effective altruist is Sam Bankman-Fried, who justified enriching himself by making risky investments with other peoples' money as a way to improve the overall welfare of the world.[3]  Bankman-Fried was to eventually give it all away in massive acts of philanthropy.  In practice, this alleged mission did not work out so well, and Bankman-Fried was sentenced last year to 25 years in prison for fraud and conspiracy related to his operation of the FTX cryptocurrency exchange.

Other well-known individuals in various rationalist circles subscribe to or are influenced by similar ideas, and have not (yet) been convicted of a crime.  Venture capitalist Marc Andreessen views rationalism and longtermism through a libertarian-accelerationist lens, viewing failure to adopt AI a greater danger than any of its existential risks.  In fact, Andreessen has provided a non-individualized list of "enemies" who purportedly stand in the way of AI growth.[4]  Andreessen's The Techno-Optimist Manifesto, a very long blog post espousing his beliefs, has a strangely messianic glow.  But no matter how you frame it, Andreessen is preaching for a low-regulation environment that would be good for Andreessen, allowing him to add multiples to the billions he already enjoys.

But Andreessen is not the only rationalist to flirt with religion.  An online community of rationalists had a major kerfuffle over the thought experiment called Roko's Basilisk.[5]  In short, it asks us to consider that a future super-intelligent AI might punish anyone who knew about it but didn't help bring it into existence.  In particular, the Basilisk would perform simulations of these people, the simulated copies subjected to extreme suffering, possibly for very long periods of time.[6]

Of course, this is just a sophisticated variation of Pascal's Wager, a much older thought experiment about the existence of God.[7]  Nonetheless, online panic ensued, leading to moderators banning discussion of the Basilisk on their forum.  And it does not end there.  According to some reports, tech CEO and armchair efficiency expert Elon Musk met the singer Grimes (mother of 3 of his at least 14 children) due to Twitter discussions of the Basilisk.[8]

And that is just the short list.  Arguments could be made that others have promoted controversial, ethically questionable, or arguably sociopathic ideas in the furtherance of AI (and often themselves at the same time).  These include OpenAI CEO Sam Altman (deemphasizing AI safety while using fear tactics to gain influence and power), investor Peter Theil (pushing for AI use in authoritarian contexts), and former Coinbase CEO Balaji Srinivasan (advocating for cloud-based governance and rule by techno-elites, with AI as a central tool).

Some AI evangelists on the more careful side of things will attempt to rationalize their arguments by stating that AGI is an engine of abundance and the distribution of the excess wealth is a political rather than technological problem.[9]  Yet, it is rare in human history for a political system to distribute wealth equally or equitably, and we have many counter-examples of the exact opposite occurring.

But let's end on an even darker note.  The Zizians are a small cult-like group that grew out of the rationalist movement.  They are currently believed to be involved in the deaths of six individuals in 2022 and 2025.[10]  To be clear, the Zizians are not mainstream rationalists.

I personally know dozens of people working in the AI space and I have to admit that they are all pretty chill.  It would not be accurate or fair to lump the majority of AI users, developers, and advocates under the "weird" moniker.

But the statements of a number of very powerful individuals, many of whom currently lead the AI community, have been more than just weird.  They are also troubling because they reveal a willingness to dismiss oversight, minimize risks, and speak in abstract, hyper-rational terms that often ignore real-world human suffering.  On the other hand, these are some of the same people who brought us aggressive online ads, social media, meme coins, and surveillance capitalism.  So their vision of the future, however well-funded and covered in altruistic gloss, deserves critical scrutiny rather than blind trust.

[1] https://en.wikipedia.org/wiki/Rationalist_community.

[2] https://en.wikipedia.org/wiki/Longtermism.

[3] https://www.newyorker.com/news/annals-of-inquiry/sam-bankman-fried-effective-altruism-and-the-question-of-complicity.

[4] Someone powerful making lists of enemies has never gone wrong.  Nope.  Not once.

[5] https://en.wikipedia.org/wiki/Roko%27s_basilisk.

[6] Ironically, the rationalists, many of whom claim to be atheists, appear to have recreated a version of the Christian hell.

[7] https://en.wikipedia.org/wiki/Pascal%27s_wager.

[8] https://www.vice.com/en/article/what-is-rokos-basilisk-elon-musk-grimes/.

[9] See, e.g., https://www.aei.org/articles/the-age-of-agi-the-upsides-and-challenges-of-superintelligence/.

[10] https://en.wikipedia.org/wiki/Zizians.

Image of Basilisco, desenho de Rodrigo Ferrarezi by Rodrigo Ferrarezi, from the Wikimedia Commons under the Creative Commons Attribution 3.0 Unported license.