Judge Albright Invalidates Patents Under Alice for the First Time

  • January 9, 2022
  • Category: Patent Litigation Feature

Western District of Texas Judge Alan D. Albright has become well known for his restrictive posture toward certain defensive motions in patent litigation, including convenience transfer motions, requests to stay litigation pending the outcome of America Invents Act (AIA) reviews, and patent eligibility challenges. While Judge Albright has granted such motions to transfer or stay on occasion, he had previously never invalidated a patent under Section 101—but that changed just as 2021 drew to a close. The invalidations capped off a year in which Judge Albright again had more patent litigation fall in his courtroom than any other judge, as noted in RPX’s fourth-quarter review.

Last month, Judge Albright issued two Alice rulings in which he invalidated the patents at issue for what appears to be the first time, granting motions filed by Facebook and Intel in two different campaigns: the first a summary judgment motion filed and decided within three months, and the second a Rule 12 motion that languished for more than a year. Beyond their inherent rarity, the decisions are each notable for other reasons: one because it ended a case just before trial, and the other for the manner in which it classifies certain Federal Circuit cases based on their characterizations of the underlying the “relevant technology”.

USC IP v. Facebook: Patent Invalidated Just Before Trial

The first of the two decisions came in litigation filed against Facebook in June 2020 (6:20-cv-00555) by USC IP Partnership, LP, which describes itself as an affiliate of Usability Sciences Corporation, “the revolutionary usability and user experience research firm founded three decades ago by IBM alumnus Jeff Schueler”. The complaint accused the company of infringing a single patent (8,645,300)—generally related to predicting the intent of a webpage visitor through a data processing engine—via its News Feed, targeting features for personalizing that service by using “signals”, “reactions”, and “stories” to “infer[] the intent of visitors”.

Before tackling patent eligibility, Judge Albright treated two other defensive motions in more typical fashion. In March 2021, he denied Facebook’s motion for intradistrict convenience transfer to the Austin Division, finding that the applicable public and private interest factors weighed “strongly” against transfer. In part, he held for the “sources of proof” and “cost of attendance for willing witnesses” factors that Facebook had not identified its supporting evidence and witnesses, respectively, with sufficient specificity. While Judge Albright ultimately concluded that the latter factor was neutral, he found that that the additional travel time to Austin required for the plaintiff’s witnesses in Dallas would materially increase witness inconvenience, applying an interpretation of the Fifth Circuit’s applicable “100-mile rule” that the Federal Circuit has previously rejected as overly rigid. That same August, Judge Albright also denied Facebook’s motion to stay the case pending the outcome of an inter partes review (IPR) against the ‘300 patent. There, he cited the fact that the final validity decision would come after the trial then scheduled for December 13, 2021; the advanced stage of the litigation; and his conclusion that the IPR would not simplify the issues for trial.

That October, Facebook then filed a motion for summary judgment under Alice, alleging that the ‘300 patent is ineligible as directed to the abstract idea of “collecting, analyzing and using intent data” at a high level (citing the claim language itself). “At the next level of abstraction”, the company further alleged, “the abstract idea can be described as providing recommended information to a visitor based on his or her current intent”. Facebook analogized the asserted claims to the search claims found unpatentable in the Federal Circuit’s decision in Electric Power Group v. Alstom and alleged that the claims “purport to address the age-old—and entirely human—problem of enabling the user to more easily locate ‘information or perform the functions that correspond to the visitor’s intent'” (emphasis in original).

In his December 20 written order granting the motion, having announced the ruling from the bench on December 6, Judge Albright found Facebook’s characterization of the claims and the comparison to Electric Power persuasive, rejecting USC IP’s argument that the claims were more closely comparable to those found eligible in DDR Holdings v. Hotels.com. While the DDR Holdings claims were found to be directed to a “technological solution to a problem unique to the Internet” involving the presentation of outside information based on a user clicking a hyperlink, Judge Albright explained, the concept claimed in the ‘300 patent of finding information to match a user’s intent “is a longstanding problem that existed long before the advent of computers and is not unique to the Internet”. As a result, he concluded that the patent is directed to an abstract idea.

Proceeding to the second step of Alice, Judge Albright then ruled that the challenged claims also lack an inventive concept, finding that they do not “contain anything ‘significantly more’ than the abstract idea itself” (citing Alice). Rather, he determined that “[t]he asserted claims recite generic and well-known ‘web pages,’ ‘web browsers’ and ‘databases’ that were well known in the prior art long before the filing of the ‘300 patent in 2011”. Considered as a whole, Judge Albright ruled that the “claims merely reflect a sequence performed in a logical order dictated by the abstract idea”. For that reason, and having rejected the plaintiff’s assertion that factual disputes precluded invalidation under the Federal Circuit’s Berkheimer decision, Judge Albright concluded that the patent is invalid under Alice. By the time the written order issued, Judge Albright had already canceled a jury trial scheduled for December 13, the eligibility determination having been announced at the hearing the week prior.

Health Discovery v. Intel: Judge Albright Laments the Tangled Web of Post-Alice Jurisprudence

Judge Albright’s second Alice invalidation came a week later, in litigation filed by publicly traded Health Discovery Corporation (HDC) against Intel in July 2020 (6:20-cv-00666) after the Patent Trial and Appeal Board (PTAB) had previously presided over a nearly decade-long interference between the two (see here for details). HDC accused Intel of infringing four pattern recognition patents (7,177,188; 7,542,959; 8,095,463; 10,402,685) through the provision of certain processors, Field Programmable Gate Arrays (FPGAs), Systems-on-Chip (SoCs), and software (i.e., those that are “tested, validated, verified, and optimized” by allegedly utilizing the claimed “Support Vector Machine-Recursive Feature Elimination” (SVM-RFE) method). As later characterized by Judge Albright, the claimed method purportedly improved upon “conventional gene selection methods” through its use of “support vector machines” (SVM) (a type of machine learning algorithm) to “identify relevant patterns in datasets” along with “recursive feature elimination” (RFE) to rank and eliminate features, thereby improving classification of datasets.

Intel filed its motion to dismiss under Alice in October 2020 under Rule 12, alleging that the four asserted patents are ineligible as directly solely to the SVM-RFE “mathematical algorithm” without the addition of “significantly more”. Rather, the company alleged that the asserted claims merely add “only computer implementation and a requirement that the analyzed information be biological”. The motion also alluded to the parties’ prior history, stating that “Intel’s motion should come as no surprise to HDC. Intel warned HDC years ago that its patent claims were directed to ineligible subject matter and that Intel did not practice the patents”.

However, Judge Albright did not decide the motion until more than a year later, on December 27, 2021, six months after issuing a ruling on claim construction. The eligibility order echoed Judge Albright’s prior commentary on the difficulty of applying Alice consistently, stating that the decision’s framework is “‘almost impossible to apply consistently and coherently’ in the context of abstract ideas” (citations omitted). “At this point, remarking on the state of eligibility jurisprudence has become hackneyed”, remarked Judge Albright. “It is nevertheless appropriate here, in an Order dismissing a complaint based on close questions arising from caselaw that is difficult to reconcile and apply”.

Judge Albright also noted how recent Federal Circuit jurisprudence has limited courts’ ability to invalidate patents under Alice in the early stages of a patent case. For instance, he noted that the Federal Circuit’s decision in MyMail v. ooVoo requires a court to resolve disputes over claim construction prior to ruling on patent eligibility (an issue not implicated here, given the court’s June 2021 order resolving the parties’ disputes over claim construction). Even more consequential is that court’s February 2018 Aatrix decision, which established that material factual disputes over a claim’s inventiveness preclude the resolution of an Alice motion at the Rule 12 stage (with the companion Berkheimer ruling applying that same rationale to summary judgment).

Yet Judge Albright spent the majority of his analysis on the broader uncertainties still posed by Alice, focusing primarily on recent Federal Circuit decisions that ask whether the claims at issue are directed to a “specific means or method that improves [that] relevant technology”. He began this inquiry by providing an overview of several key decisions in that category, separating them based on how that court has characterized “the relevant technology”, including those that find claims eligible at Alice step one as directed to technological improvements (e.g., TecSec v. Adobe (2020), Packet Intelligence v. Netscout Systems (2020), Uniloc USA v. LG Electronics (2020), and Data Engine Technologies v. Google (2018)); and those characterizing the technology “as some discrete device, system, or method”, generally finding the claims eligible (e.g., CardioNet v. InfoBionic (2020); Koninklijke KPN v. Gemalto M2M (2019), Thales Visionix v. United States (2017), and McRO v. Bandai Namco Games America (2016)). Judge Albright further noted a line of cases finding that claims satisfied the first step of Alice as being directed to a “mathematical concept, technique, analysis, or algorithm” (e.g., In re: Board of Trustees of Leland Stanford Junior University II (2021) (“Stanford II”), In re: Board of Trustees of Leland Stanford Junior University I (2021) (“Stanford I”), and SAP America v. InvestPic (2018)).

Judge Albright then proceeded to compare the claims addressed in Stanford II, SAP, CardioNet, and Koninklijke, “[i]n hopes of divining guidance” from those previous eligibility rulings. In Stanford II, Judge Albright recounted, the Federal Circuit invalidated claims related to a mathematical technique for predicting genetic information as unpatentably directed to “an enhancement to the abstract mathematical calculation, holding that the claims involved “merely an enhancement to the abstract mathematical calculation of haplotype phase itself” rather than an improvement to a technological process as in McRO and CardioNet. That same court, in SAP, invalidated claims related to “certain statistical analyses of investment information”, holding that they were abstract because they were directed to an “improvement in a mathematical technique”, not “a physical-realm improvement” of the type that saved claims in McRO and Thales. In CardioNet, the Federal Circuit overturned the Alice invalidation of claims related to a cardiac device used to reduce atrial fibrillation and atrial flutter, ruling that the claims are instead directed to a “an improved cardiac monitoring device” with various specific technological improvements detailed in the written description. Finally, in Koninklijke, the Federal Circuit overturned the invalidation of a patent at Rule 12(c), concluding that it was patentably directed to “an improved system of detecting errors during data transmission”, distinguishing prior cases invalidating claims directed to “data manipulation” as not “recit[ing] a specific enough solution to make the asserted technological improvement concrete”.

A comparison of those opinions, Judge Albright lamented, underscores the inherent challenge in extracting a “unified theory” from even this limited set of cases, let alone the Federal Circuit’s Alice jurisprudence in its entirety: “Efforts to reconcile these opinions illustrate the difficulty in applying Alice with any consistency”. For instance, he noted that CardioNet cited SAP in distinguishing between “independently abstract ideas that use computers as tools” (the claims invalidated in SAP) and those “focusing on an improvement in a technology as a tool” (the ones upheld in CardioNet). Yet both patents “select[ed] certain information, analyz[ed] it using mathematical techniques, and report[ed] or display[ed] the results of the analysis”, differing only in the type of data analyzed—and neither decisions explains where the dividing line could be found.

Also unhelpful, continued Judge Albright, is an attempt to distinguish the patents based on whether the claims related to “something physical”: while the Federal Circuit explained in SAP that the patents in McRO and Thales survived at Alice step one because they fell into that category, it ignored the tangibility of the claimed system at issue in SAP until step two. For that matter, remarked Judge Albright, it is unclear when claims directed to physical devices should be considered generic—for instance, why the device in CardioNet would be upheld but those in yet another case, Yu v. Apple, would not be. “In this Court’s opinion, the claimed components in CardioNet are not any less generic than those claimed in Yu. Yet CardioNet’s claims were deemed patent eligible; not so for the claims in Yu”. Moreover, Judge Albright argued that the type or source of data collected by a device should not render a claim more or less abstract. Relatedly, Judge Albright highlighted the inconsistency between Stanford II and Koninklijke, both of which involved data manipulation patents—but while the former invalidated claims as lacking a specific “practical application” of the claimed concept, the latter did not treat a similar “error” as dispositive.

Indeed, Judge Albright noted that different outcomes would result in the present case depending on the reasoning of the Federal Circuit cases being applied. For that reason, and “[g]iven the inconsistency riddling § 101 jurisprudence, a district court’s surest guidance rises from cases analyzing patents most like those under review”, Judge Albright concluded. “With that in mind, Stanford II is this case’s North Star, with SAP nearby”.

Applying those two holdings, Judge Albright ruled that the HDC patents here at issue are abstract under step one of Alice. “Like the patents in Stanford II and SAP, the asserted patents’ written description merely describes improving a mathematical analysis”. In both of the former two cases, the Federal Circuit “held that the claimed methods’ production of improved data relative to conventional mathematical methods could not render the claims eligible”, amounting solely to an improvement in the underlying abstract idea. That was also the case here, held Judge Albright—rendering the claims abstract at step one: “Like the claims in Stanford II and SAP, the claims here merely produce data with improved quality relative to that produced by conventional mathematical methods. (And this assumes that all the asserted claims even capture these improvements.) So, taking the allegations in HDC’s complaint as true, the asserted claims merely improve or ‘enhance’ an abstract idea. These claims, then, like the Stanford II and SAP claims, satisfy Alice’s first step”.

Judge Albright then determined that “HDC’s complaint fails to allege an inventive concept”—rejecting as unpersuasive the same allegations about the SVM-FRE method’s improvements to “data quality, such as the optimum error rate, relative to conventional methods”, that HDC made during the abstractness inquiry. Also unpersuasive were arguments related to the claimed method’s importance, broad application, and academic acceptance: “As the SAP Court held, a mathematical idea can be novel and even a ‘groundbreaking’ advance and still not be patent eligible”. Judge Albright also found that none of the additional allegations or limitations saved the claims at step two, the latter including the addition of generic components or restrictions to certain data types (which, as noted above, Judge Albright asserted should not factor into the court’s analysis).

Judge Albright then concluded by dismissing the case, though he declined to do so with prejudice as requested by the defendant.

As noted in RPX’s fourth-quarter review, Judge Albright had an eventful 2021. His substantive handling of convenience transfer motions has continued to attract the ire of the Federal Circuit, which reversed him on that issue 18 times in 2021. Meanwhile, his posture toward transfers as well as his overt efforts to seek out patent litigation to his district have led to questions from some members of the Senate—including some, like Senators Thom Tillis (R-NC) and Patrick Leahy (D-VT), who complained about both practices in a letter to Supreme Court Chief Justice John Roberts.

Check out the full report for more on Judge Albright—and for a deep dive on the various trends that shaped patent litigation and the patent marketplace in the past year.


Thank you for your feedback