The Canadian patent eligibility landscape: 18 months post-Benjamin Moore

As in many other jurisdictions, Canadian patent applications for software and other computer-implemented technologies often face subject-matter eligibility objections. For many years, the Canadian Intellectual Property Office (“CIPO”) applied internal guidance that led to a "problem-solution" style approach—similar to the European Patent Office’s inventive step analysis—to determine eligibility.

More recently, following the Federal Court’s decision in Benjamin Moore,[1] CIPO’s framework became more applicant-friendly; if a practical application could be identified for the claimed subject-matter, eligibility was more readily recognized. However, on appeal, the Benjamin Moore decision was overturned,[2] and CIPO has since reverted to a more restrictive approach. Under CIPO’s current practice, subject-matter eligibility is generally only recognized if the claimed invention results in improved functioning of the computer, or if the claims recite non-generic computer hardware.[3] In practice, this can be a high bar to meet.

In this in-depth article, we provide an overview of how the current guidance on the patentability of computer-implemented inventions has evolved, and we analyze successful (and unsuccessful) appeals to the Patent Appeal Board (“PAB”) since the FCA’s decision in Benjamin Moore.

In particular, recent PAB decisions reveal several strategies for demonstrating subject-matter eligibility, including:

1. Reciting limitations directed at guiding physical processes.
2. Demonstrating improved computer functioning or efficiency.
3. Reciting limitations directed at specialized, non-generic components.

As many software-based inventions are often performed on general-purpose computer hardware, our analysis of these PAB decisions highlights the importance of drafting claims that emphasize how the execution of software or an algorithm improves the functioning of the computer. Read on for further insights into how best to establish subject-matter eligibility before CIPO.

Table of contents

1. Statutory framework

2. How did we get here?

3. Current trends regarding subject-matter eligibility for computer-implemented inventions

4. Our recommendations

---

1. Statutory framework

Under section 2 of the Patent Act, to be patentable an invention must be directed to:

[…] any new and useful art, process, machine, manufacture or composition of matter, or any new and useful improvement in any art, process, machine, manufacture or composition of matter.[4]

In addition to fitting into one of these categories, subsection 27(8) of the Patent Act provides that no patent shall be granted for any “mere scientific principle or abstract theorem.”[5] In essence, since software (including, for example, machine learning models and other artificial intelligence solutions) lack physicality on their own, they may be considered as “mere scientific principle[s] or abstract theorem[s]” that do not fall within one of the above statutory categories.

In order to satisfy the physicality requirement, reciting a computer in the claims, in theory, is sufficient to meet the condition such that the claimed invention no longer be a mere scientific principle or abstract theorem. However, this in itself is not enough to guarantee subject-matter eligibility under CIPO’s current guidance. In particular, the current version of a CIPO Practice Notice concerning subject-matter eligibility states that:

… it is necessary to consider whether the computer cooperates together with other elements of the claimed invention and thus is part of a single actual invention and, if so, whether that actual invention has physical existence or manifests a discernible physical effect or change and relates to the manual or productive arts.[6]

The Practice Notice further provides that this physicality requirement will not be met if the “computer merely processes the algorithm in a well-known manner and the processing of the algorithm on the computer does not solve any problem in the functioning of the computer.”[7] In other words, what is required, in addition to computer implementation, is some sort of improvement in the functioning of the computer itself.

Similarly, the Canadian Manual of Patent Office Practice (“MOPOP”) states that (emphasis added):

…[a] computer program is not, by itself, statutory subject-matter. However, if the result of running the program on a computer is to provide a novel and inventive technological solution to a technological problem, then the program is viewed as modifying the technological nature of the computer as a whole. The program in such cases is not a discrete element of a claim to the computer.[8]

Critically, MOPOP states the key goal of the subject-matter eligibility analysis as identifying whether the proposed invention “provides a novel and inventive technological solution to a technological problem.”[9]

Eligibility may be met when the computer program itself is a statutory method, in other words a series of steps providing a technical solution to a technical problem, namely improving the functioning of the computer in some way. Where the software does not achieve this, CIPO will generally not consider a claim to such software to be patentable. In such cases, an alternate route to eligibility may be followed by reciting in the claim “non-generic” computer components, such as sensors or other hardware extrinsic to the computer itself.

2. How did we get here?

The current state of Canadian law on the patentability of computer-implemented inventions stems from iterative interpretations of the Patent Act by the Courts and CIPO. The FCA’s decision in Schlumberger,[10] in which mere computer-implementation was held insufficient to establish patent eligibility, is the starting point for CIPO’s current position on the patentability of computer-implemented inventions.[11]

However, in the years since Schlumberger, there has been notable inconsistency regarding the extent to which a European-style “problem and solution” approach to claims construction is to be incorporated into Canadian patent practice, with the end effect being a general restriction on the eligibility of computer-implemented inventions.

Origins

In Schlumberger, the applicant’s proposed invention was a process in which measurements for drilling boreholes in oil and gas exploration were recorded and transmitted to a computer programmed to convert the information into pre-defined formats, such as charts or graphs. The FCA was of the opinion that the novel aspects of the claims were calculations and mathematical formulae that were not patentable subject-matter, holding (emphasis added):

…If the appellant's contention were correct, it would follow that the mere fact that the use of computers is prescribed to perform the calculations prescribed in the specifications, would have the effect of transforming into patentable subject-matter what would, otherwise, be clearly not patentable. The invention of the computer would then have the unexpected result of giving a new dimension to the Patent Act by rendering patentable what, under the Act as enacted, was clearly not patentable. This, in my view, is unacceptable. I am of opinion that the fact that a computer is or should be used to implement discovery does not change the nature of that discovery.[12]

From Schlumberger onward, it became apparent to applicants that the mere inclusion of a computer in claims generally directed at non-patentable algorithms or abstract theorems would not be sufficient for CIPO to consider such claims as being patent-eligible.

The patentability of computer-implemented inventions attracted increased attention and debate in the 2010s, starting with the Amazon decision.[13] Amazon centered on a patent application for online ordering using Amazon’s “one-click” method. CIPO initially refused the application as not constituting eligible subject-matter because:

1. The application did not add anything technological in nature to human knowledge.
2. The application was directed at a mere business method.
3. There was no change in the character or condition of a physical object (i.e., the physicality requirement was not met).[14]

While the FCA ultimately ordered a redetermination of the application by CIPO, the decision confirmed the physicality requirement as extending beyond the invention merely having a practical application.[15]

Choueifaty

Patent-eligibility was later re-visited in the Choueifaty decision. At issue in Choueifaty was whether the Commissioner had applied the proper test when construing essential elements of the claims which were directed to a new method for selecting and weighing investment portfolio assets that minimized risk without impacting returns. The application had been rejected on the basis that the claims did not fall within the statutory definition of an invention.

In its determination, the PAB construed the patent claims to determine the essential elements of the claim by applying the problem-solution approach which was set out at the time in CIPO’s Manual of Patent Office Practice. Under this approach, the essential elements of the claim are those that are necessary to achieve the disclosed solution to the identified problem.

Ultimately, the PAB rejected the applicant’s arguments that the computer was an essential element of the claims, holding that the claims reflected an “optimization procedure” rather than a computer implementation that improved processing speed, and the application was refused.

The applicant appealed to Federal Court, arguing that the PAB erred in applying the problem-solution approach to construe the essential elements. The Federal Court agreed with the applicant’s argument that using the problem-solution approach to construe claims was not in keeping with the Supreme Court of Canada’s holding in Free World Trust.[16] Free World Trust had stated that:

“The words chosen by the inventor will be read in the sense the inventor is presumed to have intended, and in a way that is sympathetic to accomplishment of the inventor’s purpose expressed or implicit in the text of the claims.”[17]

Therefore, the Federal Court held that the problem-solution approach employed by CIPO at the time failed as a means to adequately purposively construe the claim.

Following Choueifaty, CIPO issued the current Practice Notice, shifting from an exclusively problem-solution approach to an approach wherein the “actual invention” is to be identified based on a purposive construction as required in Free World Trust.[18] While under this Practice Notice the problem-solution approach is no longer to be applied exclusively, the Practice Notice continues to focus on whether an “actual invention” consists of an element or elements that in cooperation provide a technological solution to a technological problem.

When an “actual invention” includes a disembodied idea, scientific principle or abstract theorem, it is only when that element cooperates with other elements of the invention that have physical existence or exert a discernible physical effect that the invention is considered patentable.

Benjamin Moore at the Federal Court and the Federal Court of Appeal

Since Choueifaty, there has been some criticism that the approach outlined in the Practice Notice continues to allow CIPO to read out the computer from claims directed at computer-implemented inventions, leading to findings that such claims were ineligible. This issue culminated in decisions by the Federal Court and the FCA involving computer-implemented paint colour selection patent applications by Benjamin Moore.

The applications had been rejected by the PAB on the grounds that the patent applications were directed to non-statutory subject-matter. Benjamin Moore argued that CIPO had once again incorrectly applied the problem-solution approach in assessing subject-matter eligibility.[19]

The Federal Court agreed with Benjamin Moore, further acknowledging comments by Benjamin Moore and the Intellectual Property Institute of Canada (“IPIC”), an intervenor, that “CIPO regularly misconstrues the patentability of computer-implemented inventions, incorrectly excluding them under section 27(8) of the Patent Act .”[20]

In light of CIPO’s inconsistent application of the Choueifaty guidance, the Federal Court adopted a new framework for assessing the patentability of computer-implemented inventions suggested by IPIC, and ordered the applications be remitted to CIPO for determination consistent with that framework.[21] The framework that IPIC proposed consisted of:

1. Purposive construction of claims.
2. Asking whether the construed claims as a whole consist only of a mere scientific principle or abstract theorem, or whether it comprises a practical application that employs a scientific principle or abstract theorem (i.e. subsection 27(8) eligibility).
3. If the construed claim comprises a practical application, assessing the construed claim for the remaining patentability criteria: statutory subject-matter categories and judicial exclusions, and then novelty, obviousness, and utility.

The Benjamin Moore framework was generally well-received by patent applicants, as the recognition of a “practical application” as a signpost for eligibility was more in keeping with the current state of software eligibility in the United States.

However, the decision of the Federal Court in Benjamin Moore was overturned by the FCA, on an appeal by the Attorney General of Canada. The FCA’s decision reflects a view that it was premature for the Federal Court to adopt IPIC’s framework on the basis that several issues regarding the patentability of computer-implemented inventions have not been properly considered by Canadian courts.[22]

The FCA noted that CIPO’s error in this case had been a failure to properly construe claims in accordance with Choueifaty.[23] Specifically, CIPO’s error had not been in applying the problem-solution approach, which is a component of the Practice Notice, but in the exclusive reliance on this approach to identify essential elements of the claims.[24]

With respect to the IPIC framework, the FCA acknowledged that the test includes the purposive construction exercise as a first step, which was in accordance with jurisprudence to date and “was never in dispute.”[25] However, the FCA held that other aspects of the framework were not adequately supported by Canadian case law, and in some ways contrary to the FCA’s decision in Amazon.[26]

Ultimately, the FCA ordered that the applications be redetermined using the current version of MOPOP, which was based on the Practice Notice issued following Choueifaty and which has been applied since this decision. However, the Court did acknowledge that subject-matter eligibility analysis for computer-implemented inventions is becoming an increasingly complex exercise in light of artificial intelligence and quantum technology, and cautioned CIPO not to “hastily conclude that the subject matter claimed is not patentable simply because it involves the use of conventional computer technology.”[27]

3. Current trends regarding subject-matter eligibility for computer-implemented inventions

The FCA’s decision to overturn the applicant-friendly Benjamin Moore framework has generally increased the bar for software eligibility. Indeed, since the FCA’s decision in Benjamin Moore, relatively few applicants have had success in challenging rejections of applications on subject-matter eligibility grounds before the PAB, as discussed below.

However, the few PAB decisions in which the applicant had success reveal important considerations for applicants in how to approach the issue of subject-matter eligibility for computer-implemented inventions.

Illustrating a physical effect

Two cases since Benjamin Moore FCA have had success before the PAB on the grounds that a demonstrable physical effect could be established. In Senior, Rodney (Re), the application was directed at a computer-implemented system comprising a database and a web server that interact to allow the purchase of fuel at retail service stations using point-of-sale equipment.[28] An independent claim recited that motor fuel could be redeemed, i.e. delivered to the consumer.

While the applicant had initially argued that the redemption was sufficient to satisfy the physicality requirement as it implied a pumping of fuel, the claims were amended to state that redemption involved a “pumping action” (i.e., to make the action of pumping explicit). The PAB found that the element “pumping action” was sufficient to render the claim patentable, whereas the claims without this element were not considered patentable.

Similarly, in Landmark Graphics Corporation (Re), the application was directed at methods and devices for prospecting, drilling, or developing oil fields and involved the input of relevant data into a computer and performing analysis of the data to guide prospecting.[29] Claims that were directed at a method of drilling and collecting data during the drilling were considered allowable, whereas a non-transitory program carrier device that was configured to model geological surfaces was found by the PAB not to be patentable subject-matter, being “generic computer components” that were used in a “well-known manner .”[30]

Both these decisions show that limiting the claims to physical effects or actions can make a claim patent-eligible. However, this approach generally requires more than arguing that the subject-matter exerts a real-world effect. In recent decisions, the PAB has emphasized the holding of the FCA in Amazon that the requirement for something with physical existence or that exerts a discernible physical effect or change cannot be met merely by the fact the claimed invention has a practical application.[31]

For instance, an application directed at methods for nutritional planning that interacted with point-of-sale devices, similar to Senior, Rodney (Re), was found not to relate to patentable subject-matter.[32] In that case, the actual invention was found to be limited to an abstract nutritional planning algorithm lacking physical existence.[33] It appears that claim limitations in successful applications must go beyond improvements in disembodied processes, such as planning, to show a specific physical effect, such as control of a physical process such as drilling.

Improving efficiency or computer functioning

The Practice Notice generally makes it clear that a claim that results in improved functioning of the computer will render the claim patentable:

On the other hand, if running the algorithm on the computer improves the functioning of the computer, then the computer and the algorithm would together form a single actual invention that solves a problem related to the manual or productive arts and the subject-matter defined by the claim would be patentable subject-matter and not be prohibited under subsection 27(8) of the Patent Act.

CIPO is generally interpreting the improved functioning of the computer in a restrictive sense. Often, the applicant must show that the computer on which the claimed software is being executed is using fewer resources, or is operating with increased efficiency, when compared, for example, to the prior art.

In Creditex Group, Inc (Re), the application was directed at methods and platforms for post-trade affirmation and a messaging service for credit derivative trading.[34] The PAB queried whether these elements were part of a single actual invention with the processing algorithm underlying the trading. While the applicant argued that their invention improved the efficiency of the overall system by pre-reconciling trade data before it was processed by a downstream trade processing system, this argument was initially rejected on the grounds that pre-reconciling of the data is merely a shift of processing to a different part of the overall processing pipeline, rather than an improvement to the efficiency of the overall system.

However, before the PAB, the applicant argued that, since the downstream processing could be an existing third-party system, providing such a system with pre-reconciled data eliminated the need for reconciliation checks, in turn improving the overall system’s efficiency. The PAB accepted this argument, holding:

“The downstream computer system would be part of an actual invention manifesting a discernible physical effect in terms of its processing capacity, even though the effect is caused by an algorithm executed by computerized element in another part of the system.”[35]

In contrast, BGC Partners (Re) is an example of an unsuccessful attempt to argue for improvements to functioning of the computer.[36] The application in question was directed at systems and methods that apply rules to execute market orders, such as whether to buy or sell a stock at the current market price.

The applicant argued that features such as receiving information over a communication network, real-time reception of information, and automatic determinations of market stability and changes represent improvements in functioning of the computer: by controlling network activity and computer workload, the applicant argued the computer was able to perform functions it was previously not able to perform.

The PAB rejected this argument on the grounds that there was no basis in the description for these alleged benefits. Further, the PAB found that benefits such as high-speed transfer and processing of information were expected benefits of computer systems in general.[37] Interestingly, the enabling of new functionality argued by the applicant was viewed by the PAB as insufficient to ensure patentability if this new functionality came only from the processing of an algorithm in a well-known manner.[38]

Collectively, these two decisions underscore CIPO’s view that arguments related to improvements in computer efficiency and functionality must be related to an identifiable improvement that is supported by the description of the application.

Specialized components

A final route to demonstrating subject-matter eligibility is through the inclusion of specialized computing components, as illustrated in Brink’s Network, Inc (Re).[39] This application was directed at a computing system comprising a specialized safe. The PAB noted that the safe was a specialized safe comprising components not part of a generic computer system.

Since these components cooperated with functions of the safe, such as transmission of information over a network, they were part of the actual invention and met the physicality requirement. While not feasible for all applicants for patents on computer-implemented inventions, Brink’s reveals the importance of emphasizing how specialized components cooperate with the invention’s overall computing functions to show subject-matter eligibility.

4. Our recommendations

Our analysis of recent PAB decisions reveals that, although applications for computer-implemented inventions continue to face difficulty in a post-Benjamin Moore landscape, there are viable approaches for showing subject-matter eligibility.

In particular, when drafting an application for computer-implemented technology, is it important, to whatever extent possible, to include any discussion of improved functioning of the computer as a result of execution of the software. For example, discussion of decreased memory requirements, increased processing speed, or reduced processing times, should be included, together with a description of how these changes in computer functioning are implemented.

Such discussion may be made with comparison to the prior art. While it is possible that submitting evidence post-filing of such improved functioning of the computer may be allowed (for example, via an Affidavit or a similar declaration), an explicit reference to such improved functioning in the patent specification itself may carry more weight.

Applicants may even consider running side-by-side tests of their claimed software against incumbent solutions and including the results of any such tests in the application. Any indication that the solution provided by the computer-implemented technology can be arrived at in a more computer-efficient manner than the prior art may assist with any downstream eligibility objections that are raised by CIPO or even a third party.

As the Creditex decision shows, even where a specific description of improved efficiency is missing, it may be possible to demonstrate, through explanation of how components of the invention cooperate, that the functionality of the invention’s computer is improved relative to existing technology, illustrating one viable approach to addressing subject-matter rejections.