Evolution of the patenting of computer-related inventions in India
The rise of AI has intensified the global focus on the protection of computer-related inventions. Pankaj Soni and Neha Malhotra of Remfry & Sagar explain how India’s legislature and judiciary have addressed the issue
With the buzz around ChatGPT and similar tools, artificial intelligence (AI) has become the hot topic of conversation everywhere. While the world debates the disruptive reach of these tools and, consequently, the impact on life as we know it, the innovator community – governments and individuals – is trying to understand the parameters of intellectual property (IP) protection available for such tools and inventions. But, before we get there, the protection of computer-related inventions (CRIs), in some ways the foundation of AI inventions, becomes more critical.
The jurisprudence for the patentability of CRIs has been debated for many years across all major patent jurisdictions, and in light of AI tools, they are now being looked at even more closely across the globe.
“Computer programme per se”
In India, “a mathematical or business method or a computer programme [sic] per se or algorithms” are excluded from patentability under the provisions of Section 3(k) of The Patents Act, 1970 (the Act). The phrase “computer programme per se” in Section 3(k) was introduced in The Patents (Amendment) Act, 2002, based on the recommendation of the Joint Parliamentary Committee (JPC) Report, 2001. The addition of “per se” to “computer programme” shows the clear legislative intent to grant patents to inventions that go beyond being mere computer programs.
Support for this proposition is found in the JPC report, which clarified that “[t]his change has been proposed because sometimes the computer programme may include certain other things, ancillary thereto or developed thereon. The intention here is not to reject them for grant of patent if they are inventions. However, the computer programmes as such are not intended to be granted patent.”
Although the intent of the JPC to include per se is seemingly clear, patent practice has evolved in a somewhat arbitrary and non-uniform manner. The primary reasons are the lack of a proper definition of the phrase “computer programme per se” in the Act, and the ever-changing CRI guidelines of the Indian Patent Office (the Patent Office), which did not explain what interplay between hardware features and technical aspects of the invention would need to be satisfied to allow for the patentability of CRIs.
A major point of contention has been (by some controllers) the insistence on including novel hardware features in claims relating to CRIs. This approach has no basis in the Act or The Patents Rules, 2003 and has resulted in a headache for many applicants.
To make things more confusing for applicants, judicial decisions over the years have ruled against the restrictive approach of the Patent Office. Starting with the erstwhile Intellectual Property Appellate Board’s (IPAB’s) decisions in Enercon India Ltd. v Aloys Wobben (2010) and Accenture Global Service GMBH v The Assistant Controller of Patents & Designs (2012), case law has shed some light on the patentability of CRIs and offered a way to separate patentable computer-related applications from the ambit of “computer programme per se”.
In the Accenture case, the patent application relating to a system and method for developing an internet-hosted business application composed of web services was initially refused by the Patent Office on the grounds of Section 3(k). The Patent Office relied on a twofold standard for analysing the patentability of the invention to ascertain whether:
The invention related to a hardware implementation performing a novel function; and
The novel features of the invention resided in a set of instructions designed to cause the hardware to perform the desired operations without special adoption of the hardware, or the novel features of the invention resided in modification of the hardware (novel hardware).
The Patent Office concluded that in the absence of novel hardware, the invention was precluded from patentability under Section 3(k).
In an appeal before the IPAB, Accenture asserted that the standards relied upon by the Patent Office are not prescribed in the Act or the Manual of Patent Office Practice and Procedure. The IPAB agreed, and observed that the ‘novel hardware’ feature is not mandatory to clear the bar of Section 3(k). The application was remanded back to the Patent Office, which later granted the application and found that the invention resulted in the improvement of web services and software, thereby not falling under the category of “computer programme per se” as given in Section 3(k) of the Act.
Following this decision, the Patent Office endeavoured to streamline its examination practices through its guidelines and manuals, and issued its draft Guidelines for Examination of Computer Related Inventions (CRIs) in 2013. The guidelines defined the terms “technical contribution” and “technical effect” as valid criteria for patentability and laid down the parameters based on which CRIs should be examined by the controllers. The guidelines also provided illustrations and examples of eligible and non-eligible inventions.
However, the fact that the guidelines were not legally binding – and with no clear definition of the phrases “per se”, “technical advancement” and “technical contribution” in the Act – meant the Patent Office’s position varied based on the view of the individual controller examining the application and the position on the patentability of CRIs under the Indian patent regime remained ambiguous for stakeholders.
Although the need for streamlining the law for patenting CRIs in India continued to be felt strongly, not much happened until 2015–16, when Section 3(k) again garnered attention in the Ericsson standard essential patent saga (against Intex Technologies and Lava International Ltd.), wherein the validity of Ericsson’s patents was challenged on the grounds of Section 3(k).
In these cases, the High Court of Delhi observed that inventions exhibiting a technical improvement and capable of actual physical representation are not excluded under Section 3(k). Particularly, in the Intex case, the Single Bench opined that an “invention having technical contribution or technical effect is not merely a computer programme per se and should be patentable”. These findings were reaffirmed on March 29 2023, by the Division Bench of the High Court of Delhi.
Impelled by judicial precedents and the rapidly growing technology space relating to CRIs, the Patent Office revised its CRI guidelines in 2016. Unfortunately, and to everyone’s surprise, the 2016 guidelines were more restrictive compared to the 2013 version.
After receiving strong criticism from the stakeholders, the 2016 guidelines were revised in 2017 and like the earlier iteration of the guidelines and judicial precedents, the 2017 version attempted to demystify the ambiguity surrounding Section 3(k) by establishing a relationship between a “computer programme per se” and technical advancement/contribution. The new paradigm stated that inventions featuring technical advancement or exhibiting a technical effect would not fall under the category of “computer programme per se” and were eligible for patent protection.
This position was affirmed by the High Court of Delhi in Ferid Allani v Union of India & Ors. (2019), wherein the court allowed a writ petition against the order of the IPAB, and held that Section 3(k) only bars inventions relating to a “computer programme per se” and not all computer programs. The court observed that if the invention exhibits a technical effect and a technical contribution, it is said to meet the eligibility criteria for patentability.
Current position – technical effect is determinative
Despite the judgments and revisions in the guidelines, the current position on the patentability of CRIs in India continues to suffer the residual effects of inconsistent application by the Patent Office, which, unsurprisingly, has led to the erroneous rejection of deserving applications. When coupled with unreasoned rejection orders – i.e., decisions by the Patent Office that give no analysis and/or lack detailed reasoning – the situation puts the applicant in a predicament because they have no basis of knowing the position the Patent Office is willing to take with respect to CRI patentability.
This issue was recently discussed in the High Court of Delhi’s judgment in Microsoft Technology Licensing, LLC v Assistant Controller of Patents and Designs (May 15 2023).
The petitioner, Microsoft, had filed an application relating to a two-tier authentication by way of two different cookies. The claimed technique secures the authentication of a user while accessing one or more sub-location(s) in a network address and effectively foils the attempts of a malicious user to gain access to network sub-location(s) by illegally obtaining cookies from another user.
The controller, in his refusal decision, held that the subject matter of the claims relates to a “computer programme per se” and the inventive step of the invention lies in the non-patentable subject matter under Section 3(k) of the Act.
Microsoft assailed the decision of the controller; inter alia, on the ground that the controller’s interpretation of Section 3(k) was incorrect. Relying on the precedents of the court in the Ferid Allani and Intex cases, Microsoft argued that the invention cannot be categorised as a “computer programme per se” as it relates to a technical process and solves a technical problem of securing data accessed on a network.
The Patent Office opposed the appeal by arguing that Microsoft had failed to appreciate the legislative intent behind the Section 3(k) provision and the controller rightly rejected the application as the invention prima facie lies in the set of instructions implemented on a “computer programme per se”.
The High Court of Delhi renders its decision
The High Court of Delhi comprehensively discussed:
The origin and evolution of Section 3(k) of the Act;
The legislative intent behind the inclusion of “per se” in the Act; and
The CRI guidelines of the Patent Office.
It also discussed the tests of technical effect and reviewed the legal positions in the US and the EU on the patentability of CRIs.
Emphasising the need to adopt a clear definition of per se for proper application of the law, the court held that “per se” makes it clear that “computer programmes as such” are non-patentable. Accordingly, inventions which exhibit a technical effect and provide a technical solution to a technical problem though an underlying software may be patentable.
Reaffirming the point of law in the Ferid Allani case, the court added that a proper understanding and application of technical advancement and technical effect are crucial in any determination of the patentability of CRIs.
The court clarified that technical effect or contribution can be shown in an invention if the invention solves a “technical problem, enhances a technical process or has some other technical benefit”. It held that just because an invention merely involves a mathematical method or computer program per se, it does not render the invention to be non-patentable under the Act.
Expressing discontent with the Patent Office’s stringent requirement of novel hardware for allowing a CRI application, the court held that “patent applications should be considered in [the] context of established judicial precedents, Section 3(k) provisions and extant guidelines related to CRIs and other materials indicating [a] legislative framework”.
The court also observed that the examiners should analyse each application individually, understanding the core of an invention, rather than focusing solely on the claims of the invention, to allow patent protection for deserving applications that feature technical advancement and offer a solution to real-world problems.
Criticising the deletion of examples and illustrations of patentable and non-patentable inventions from the 2017 CRI guidelines, the court also directed the Patent Office to prepare ‘signposts’ and ‘indicators’ based on case law to the examiners, which may provide guidance for examination and further help in achieving uniformity in the examination process of CRIs by the Patent Office and consistency with standard international practice, such as that of the USPTO and the EPO.
In the end, the court found that the claimed invention offers a novel and inventive solution to a security problem by providing a two-tier authentication process and improving the overall user experience. The court found the controller had overlooked the technical aspects of the invention; it remanded the matter back to the Patent Office and asked it to reconsider the application based on the criteria subject to which a computer-related invention must be analysed as a whole, including its technical contribution.
A critical moment for the Patent Office
In a report published by Nasscom in April 2023, more than 50% of the emerging technology patents filed in India over the past decade were from internet of things or AI domains. The fast-paced growth of software technology therefore calls for the creation of a sustainable IP ecosystem for technology-driven inventions, with relaxed laws for patenting and protecting software technologies in India.
2023 began with judgments from the High Court of Delhi on the proper application of Section 3(k). In addition to the Microsoft judgment, on March 29 2023, the Division Bench of the High Court of Delhi, in Intex v Ericsson, reaffirmed the findings of the Single Bench that an “invention having technical contribution or technical effect is not merely a computer programme per se and should be patentable”.
In OpenTV Inc. v The Controller of Patents and Designs and Anr, the court held that inventions that have a technical effect and solve a technical problem can be excluded from the scope of Section 3(k). The court also referred to the 161st Report of the Review of the Intellectual Property Rights Regime in India presented by the parliamentary standing committee, and opined that “a large number of inventions in emerging technologies including by SMEs, start-ups and educational institutions could be in the field of business methods or application of computing and digital technologies, therefore, there is a need to have a re-look at the exclusions in Section 3(k) of the Act, in view of the growing innovations”.
While the recent judgments are a step forward, what awaits us is how the Patent Office will react and adopt new practices. Principally, a more well-balanced and consistent approach, which has proved elusive for applicants over the years, is required by the Patent Office to allow worthy CRI owners to seek patent protection in India and help the IP regime to reach its true potential.