Since the U.S. Supreme Court's 2014 decision in Alice v. CLS Bank, which effectively put the brakes on securing software patents in the US, patent practitioners have been working to find ways to overcome the bar Alice established for satisfying the subject matter eligibility requirements of 35 U.S.C. Section 101. Alice was focused on whether claims for software and business methods were eligible subject matter under Section 101. The Alice court held that merely describing an algorithm at a high level and having that algorithm execute on a generic computing device constitutes an "abstract idea," and is not eligible for patent protection.

Over the years since Alice, there have been a number of evolutions to the standard by which applications were evaluated. Initially, courts and examiners took a heavy hand in invalidating or rejecting claims drawn to software and business method patents. For example, examiners frequently relied on Electric Power Group v. Alstom, which held that the process of collecting information, analyzing it, and displaying certain results of the collection and analysis constitutes an abstract idea, to render the claims unpatentable under Section 101. There were, however, a limited number of cases that helped ease the restrictions set forth by Alice. In Enfish v. Microsoft, the U.S. Court of Appeals for the Federal Circuit upheld the eligibility of the claims at issue, which were directed to an improved database structure, finding that they recited an improvement in the functioning of the computer itself and thus constitutes an invention under Section 101. In BASCOM Global Internet Services v. AT&T Mobility, the Federal Circuit concluded that the claims directed to a content filtering system satisfied the requirements under Section 101, despite the fact that the claims were directed to an abstract idea, because an inventive concept could be found in an ordered combination of the claim elements. In Core Wireless Licensing S.A.R.L. v. LG Electronics, the Federal Circuit found that claims reciting a display screen to be patent eligible because the claims described a specific implementation of how information is presented or displayed to a user via a graphical user interface. As the courts' and U.S. Patent Office's standards for patenting software and business methods have eased since Alice, new concerns with the patentability of these claims have arisen. Namely, as the landscape of eligible subject matter under Section 101 has become clearer, courts and examiners are now turning their attention to patentability under 35 U.S.C. Section 112, which governs the standard by which the written description of the invention is evaluated.

In January 2019, the U.S. Patent and Trademark Office (USPTO) published the "2019 revised patent subject matter eligibility guidance" (2019 guidance), which clarified new guidelines for subject matter eligibility under 35 U.S.C. Section 101. The revised examination procedures were "designed to more accurately and consistently identify claims that recite a practical application of a judicial exception (and thus are not 'directed to' a judicial exception), thereby increasing predictability and consistency in the patent eligibility analysis." Following the 2019 guidance, the USPTO further published the "October 2019 update: subject matter eligibility" (October update) that provided a more in depth discussion to the various guidelines set forth in the 2019 guidance. As a result of the 2019 guidance and the October update, heightened bar for overcoming subject matter eligibility challenges has lowered. In fact, according to a recent article on IPWatchdog, in 2019 61.8% of issued U.S. utility patents were "software-related;" in 2018, only 21.64% of issued U.S. utility patents were "software-related."

Simultaneous with the release of the 2019 guidance, the USPTO also released guidance directed to computer-implemented functional claim limitations for compliance with Section 112. In "Examining Computer-Implemented Functional Claim Limitations for Compliance with 35 U.S.C. 112" (Section 112 Guidance), the USPTO addressed issues relating to the software claims having means-plus-function limitations as well as issues related to written description and enablement for these claims. The Section 112 guidance sets forth elevated standards for applicants to secure patent protection on computer-implemented methods. One likely effect of the Section 112 guidance is that examiners will have an increased focus on the requirements for written description and enablement (Section 112(a)), definiteness (Section 112(b)), and functional limitation challenges (Section 112(f)). Under Section 112(a), the specification must contain a written description of the invention such that one skilled in the art can reasonably conclude that the inventor had possession of the claimed invention. Under Section 112(b), the claims must adequately define the metes and bounds of the subject matter. Under Section 112(f), a claim may be expressed as a means for performing a specified function without a recital of structure, but that structure must be adequately described in the detailed description. Below are some guidelines and considerations for preparing and prosecuting software patents to navigate the restrictions under Section 112.

Functional Claim Limitations Under 35 U.S.C. Section 112(f)

If a claim limitation recites a term and contains functional language associated with the term,  USPTO examiners are advised to determine whether the claim limitation invokes 35 U.S.C. Section 112(f), which requires the detailed description to have an adequate description of the structure corresponding to the functional limitation. The USPTO has set a 3-prong analysis for determining if a claim limitation invokes Section 112(f). At a high level, the Section 112 Guidance describes the three-prong analysis as evaluating whether: the claim limitation uses the term "means," "step," or a generic placeholder; the term is modified by functional language; and the term is not modified by sufficient structure, material, or acts for performing the function. Examiners are further advised to determine whether the claim sets forth sufficient structure for performing the claim limitations. Examples of structure may include the components of the computer system that carry out the claimed limitation, such as, but not limited to, a processor, memory, network interface card, integrated circuit, input/output devices, a processor implementing a certain algorithm, etc. If an examiner can demonstrate that the claims meet each of the three-prongs in this analysis, the burden may then switch to the applicant to: demonstrate that the claim limitations recite sufficient structure to perform the claimed function so to as to avoid scrutiny under Section 112(f); or amend the claim limitation in a way that avoids interpretation under Section 112, such as by omitting a generic placeholder or sufficiently describing the structure performing the claim limitation.

Indefiniteness Under 35 U.S.C. Section 112(b)

Assuming that a claim limitation invokes Section 112(f), in order to avoid an indefinite challenge, the specification must disclose and describe an algorithm for performing the claimed specific computer functions in accordance with Section 112(b). To rebut an indefiniteness challenge, it is insufficient for applicants to merely argue that one of ordinary skill in the art would be capable of writing software to transform a general purpose computer into a special purpose computer programmed to perform the claimed operation. Rather, the USPTO advises that the sufficiency of the algorithm is determined in view of what one or ordinary skill in the art would understand as sufficient to define the structure and make the boundaries of the claim understandable. For example, assuming that a special purpose computer is programmed to perform three specialized functions, the specification must provide enough detail for carrying out each of the three specialized functions. If, for example, the specification fails to describe an algorithm for such a failure would render the claim term indefinite.

Written Description Requirement and Enablement Under 35 U.S.C. Section 112(a)

In the Section 112 Guidance, the USPTO advised that the test for whether a specification complies with the written description requirement calls for a comparison between the scope of a claim with the scope of the description supporting the claim to determine whether the applicant has demonstrated possession of the claimed invention. In other words, the function of the computer system must be described in sufficient detail to describe how the system works, rather than simply pointing to black box type aspects of a system that perform a certain function. For computer-implemented functional claims, this inquiry is performed on both the disclosed hardware and disclosed software, due to the interrelationship and independence of the hardware and software. For the enablement requirement of Section 112(a), Examiners are advised to determine whether the specification sufficiently teaches those skilled in the art how to make and use the full scope of the claimed invention without undue experimentation.

What This Means for Applicants

From the Section 112 Guidance one issue remains clear: now more than ever, applicants must sufficiently describe how a claimed function is achieved and not merely the effect of a claimed function.  For computer-implemented limitations, applicants can satisfy this burden by sufficiently describing the algorithm underlying the claim limitations (including inputs, analysis steps and output) and any hardware that is necessary for carrying out that algorithm. Based on Section 112 guidance, claims that rely on various "modules" for carrying out functions, without providing sufficient details of what happens at or within each module, are likely to be considered invalid. The Section 112 guidance treat such "black box" structures as if the applicant used the term "means" in describing the functions carried out by the module.

A particular area where the new emphasis on evaluating claims under Section 112 may take a more prevalent role in affecting patentability is in machine learning based inventions. Although the 2019 guidance and October update have provided applicants with sufficient avenues for overcoming subject matter based challenges in this technical area, the Section 112 guidance places a heightened burden on Applicants to ensure that the modules or components recited in the claim are sufficiently described in the detailed description. For example, a "black box" machine learning system that merely receives an input and outputs a result or prediction would be subject to scrutiny under Section 112(a) and Section 112(b) (e.g., receiving a data set, inputting a data set into a "neural network," or "machine learning module," or "prediction engine," etc., and outputting a result based on the data set). To proactively avoid a Section 112(a) or Section 112(b) challenge, applicants should sufficiently describe the components of the machine learning system and the associated functions. Exemplary details in the specification that could aid in overcoming or anticipating a Section 112(a) or Section 112(b) challenge may be: a sufficient description of the training and testing data sets, the type of machine learning model being used, and hyperparameters of the module. Similarly, for inventions that are not based on machine learning, a description of a module's effect instead of how the module achieves that affect would be vulnerable to Section 112(a) or Section 112(b) challenges.

Although the USPTO has continued to provide helpful guidelines for overcoming Section 101 challenges, the Section 112 guidance that silently accompanied the 2019 guidance may create additional barriers to patentability for applicants. If the past five years of developments since the Alice decision are any indication, the standards for patentability under Section 112 are likely to continue to evolve as new cases evaluating patent validity are decided and the USPTO attempts to have its Examiners evaluate applications consistent with judicial standards for validity.

William Bartow is a partner in DLA Piper's Philadelphia office. His practices focuses on intellectual property law, including patents, trademarks, copyrights, licensing and litigation.

Joseph Wolfe is an associate in the Philadelphia office of the firm. He concentrates his practice on intellectual property, including domestic and foreign patent prosecution and preparation, patent validity and infringement analysis, and due diligence.