Breaking Down the Draft CRI Guidelines 2025: A Critical Review for Tech Innovators and IP Professionals

4th April, 2025 - 7:31 am Categories: CRI Guidelines 2025 0 Comments

A Closer Look at the Draft CRI Guidelines

In an era defined by rapid technological innovation, the patent examination process must evolve to accommodate advancements in information technology. The Draft CRI Guidelines 2025 (hereinafter “Guidelines”), published by the office of the CGPDTM on March 25, 2025, aim to clarify the patentability criteria for Computer-Related Inventions (CRIs) in India.[i] While these Guidelines claim to standardize and bring greater transparency to the patent examination process, a closer examination reveals areas of ambiguity, potential loopholes, and concerns regarding their practical implementation.

Key Definitions: Demystifying or Creating New Ambiguities?

A critical component of the Guidelines is the definition of key terms that play a role in determining patentability. However, while the Guidelines attempt to demystify several technical and legal concepts, they may not fully resolve longstanding uncertainties. Key terms include:

• Algorithm and Computer Programme per se: The Guidelines reiterate judicial interpretations, stating that algorithms and computer programs per se are excluded from patentability unless they demonstrate a “technical effect”. However, the reliance on Black’s Law Dictionary (cited in Microsoft Technology Licensing LLC v. Assistant Controller of Patents And Designs, 2024)  to define “per se” does not address the lack of clarity surrounding what qualifies as an “additional technical effect”. What constitutes a genuine technical contribution remains a matter of interpretation and could lead to inconsistent examination outcomes. The Guidelines also do not clarify whether AI-generated algorithms with dynamic learning capabilities would be treated differently from static algorithms.
• Computer and Computer Network: The Guidelines align these definitions with the Information Technology Act, ensuring that any invention involving these elements is evaluated on a clear technical foundation. However, given that CRIs frequently involve distributed computing, cloud-based architectures, and AI-driven processing, these definitions may be too narrow to accommodate modern technological realities. In other words, the Guidelines fail to explicitly clarify how examiners should assess software-implemented inventions that rely on network-based functionality, especially in decentralized computing environments such as blockchain-based applications.
• Data, Hardware and Secure Systems: By attempting to set standardized definitions, the Guidelines seek to eliminate ambiguity, but doing the same has instead introduced new areas of contention, particularly in addressing the challenges that persist in determining the extent to which software-driven innovations qualify as technical advancements. Additionally, the Guidelines do not offer a clear stance on the treatment of data-driven innovations, including big data analytics and AI-generated insights, under the existing patent framework. Given the pace at which computing architectures evolve, the risk remains that these definitions could become outdated or misapplied in future cases.

The Legal Framework: Refining or Repeating Existing Challenges?

The Guidelines delve into the evolving legal landscape by reviewing amendments to the Patents Act and recent court decisions. They address key statutory exclusions, particularly under Section 3(k) of the Patents Act, which excludes “mathematical methods, business methods, algorithms, and computer programmes per se” from patentability. While the Guidelines clarify that an invention demonstrating a genuine technical contribution may overcome these exclusions, they do not offer a precise methodology for distinguishing between a mere implementation of a software-based idea and a truly patentable technical innovation.

According to the Guidelines, recent court judgments have played a pivotal role in shaping these interpretations. However, the varying interpretations of “technical effect” across different cases create uncertainty in application.

• Technical Effect and Contribution: By placing reliance on various cases, the Guidelines state that a CRI must provide a technical effect or solve a technical problem, rather than merely implementing a computer program. However, the Guidelines do not define clear parameters for assessing technical contribution beyond broad principles, leaving room for subjectivity. Further, the examples provided are limited in scope and fail to account for edge cases, such as AI models that enhance efficiency or autonomously optimize performance, but do not introduce a fundamentally new computational method. It remains unclear as to how such advancements will be examined under existing frameworks.
• Exclusions Revisited: Judgments concerning business methods and mathematical algorithms have reinforced that while these elements may be inherent to a broader invention, they cannot be the sole basis for a patent claim. While the Guidelines underscore that the substance of the invention—its overall technical merits—must be the primary focus, the criteria to determine when an invention “merely” automates a business process is unclear. Thus, there is a risk that vague claim language may result in subjective interpretations by patent examiners, potentially leading to inconsistent grant or rejection decisions.

Examination Procedures: Do They Truly Standardize the Process?

While the Guidelines introduce structured methodologies for evaluating CRI patent applications, their effectiveness in practical examination remains to be seen.

• Novelty and Inventive Step: The “Seven Stambhas Approach” for assessing novelty[ii] and the “five-step test” for assessing inventive step[iii] aim to bring consistency to patent examination. However, their effectiveness hinges on examiner training and strict adherence to objective evaluation criteria. There is still no clear guidance on how an invention’s “contribution” should be assessed when combining known techniques in novel ways, a common scenario in software patents. Further, the five-step test for assessing inventive step offers a subjective assessment criteria that may lead to inconsistent interpretations across applications. The Guidelines also do not provide clarity on how incremental but significant improvements—such as AI-driven efficiency enhancements—would be assessed under this framework.
• Sufficiency of Disclosure: A Heightened Standard?: The Draft CRI Guidelines 2025 reinforce the principle that patent protection is granted in exchange for complete and transparent disclosure. They emphasize that Computer-Related Invention (CRI) applications must sufficiently describe both the what (the invention’s structure and components) and the how (the best method of performing it) to enable a skilled person to replicate the invention without undue experimentation.

(a) AI-Based Inventions: The Guidelines explicitly require AI-related patent applications to disclose specifics such as training data sources, model architectures (e.g., neural network layers), and learning algorithms. This aligns with case law, as seen in Caleb Suresh Motupalli v. Controller of Patents (2025), where a patent was rejected for failing to explain the functionality of a “black-box modernization technique.” However, this requirement raises concerns for applicants who may seek to protect trade secrets related to proprietary AI models. The Guidelines do not address whether partial disclosure (e.g., high-level functional descriptions) would suffice, leaving uncertainty regarding the balance between disclosure and confidentiality.

(b) Blockchain Applications: The Guidelines mandate the disclosure of cryptographic techniques, consensus mechanisms, and data structures, ensuring that blockchain-related inventions meet the same enablement standards as other CRIs. However, they remain unclear on whether modifications to existing blockchain protocols—such as efficiency improvements or novel consensus variations—constitute a patentable technical advancement. Without clear distinctions, applicants may face challenges in demonstrating novelty and inventive step for blockchain-based innovations.

Determination of Excluded Subject Matter: A Loophole for “Technical Effect”?

A major shortcoming of the Guidelines is their potential to create inconsistencies in determining exclusions under Section 3(k).

• The Guidelines emphasize that an invention must demonstrate a “tangible technical effect,” but they provide no objective test to measure this effect.
• Merely implementing an abstract idea on a computer does not make it patentable. The invention must offer a technical solution to a technical problem, such as improving computer efficiency, data storage, or network security. For instance, a generic sorting algorithm is not patentable, but an algorithm that enhances database search speed through a novel indexing mechanism may be considered. In essence, CRI patents may be granted for technological advancements, rather than mere abstract ideas, ensuring that patent law protects true innovations instead of fundamental concepts. However, the lack of a standardized approach for distinguishing genuine technological advancements from clever drafting techniques aimed at circumventing exclusions creates uncertainty. This leaves room for subjective interpretation by patent examiners, potentially leading to inconsistent application of the Guidelines.
• The Guidelines though emphasize on the importance of technical effect and technical advancement but do not specify clear criteria for evaluating improvements in computing efficiency (e.g., novel data compression methods). This omission leaves significant room for examiner discretion, increasing the risk of inconsistent or arbitrary decisions.

Illustrative Examples: Are They Sufficiently Representative?

While the Guidelines attempt to bridge the gap between theory and practice through illustrative examples, they remain limited in scope. For instance:

• The example of a two-cookie authentication system illustrates how a technical solution improves security, but the Guidelines do not clarify how patent examiners should evaluate or treat incremental improvements to existing security protocols, potentially leaving room for subjective interpretation.
• The Guidelines fail to provide clear examples for AI-generated inventions, where the role of human intervention in algorithm design remains contentious.
• The focus remains on software-hardware integration, leaving uncertainties for purely software-driven innovations.

Implications for Stakeholders: Does It Truly Improve Clarity?

The Guidelines have far-reaching implications for patent examiners, applicants, and legal professionals. However, whether they successfully improve clarity remains debatable.

• Patent Examiners: While the structured approaches provide a framework, they do not eliminate subjective interpretation. Without further training, inconsistencies in examination practices are likely to persist.
• Applicants: The Guidelines provide clearer expectations, but they also raise new challenges, particularly for AI and blockchain patents, where the sufficiency of disclosure requirements may conflict with proprietary business interests.
• Legal Professionals: While the Guidelines attempt to consolidate judicial interpretations, ambiguities remain regarding how examiners should assess the “technical effect” of CRIs, which may lead to continued litigation over patent rejections.

Conclusion: Do the Guidelines Solve Past Issues or Create New Ones?

The Draft CRI Guidelines 2025 represent a much-needed attempt to bring consistency to CRI patent examinations. However, the absence of well-defined criteria for assessing technical effect, the potential for subjective application, and the lack of practical examples limit their effectiveness. The sufficiency of disclosure requirements for AI and blockchain patents create added challenges for applicants balancing patent protection with trade secrecy. As stakeholders provide feedback during the consultation phase, it is imperative that these concerns are addressed to ensure that the Guidelines truly support technological innovation while maintaining legal rigor.

The coming months will determine whether the Guidelines serve as a progressive step forward or if they simply reinforce existing ambiguities under a different framework. Until then, innovators and IP professionals must tread carefully in navigating India’s evolving CRI patent landscape.

What needs to be clarified?

1. Clearer Definition of “Technical Effect” – A consistent test should be provided to avoid arbitrary rejections.
2. Refinement of the 7-Step Novelty Test – The test should account for software innovations rather than favouring hardware-based solutions.
3. Guidance on AI and Emerging Tech – A specific framework should be introduced for assessing AI, blockchain, and quantum computing patents.
4. Clarification of Section 3(k) Exclusion – A well-defined boundary between “software per se” and “software with technical contribution” is essential.

Author- Kuldeep Singh, Principal Associate at Khurana & Khurana, in case of any queries please contact/write back to us via email to chhavi@khuranaandkhurana.com or at Khurana & Khurana, Advocates and IP Attorney.

[i]Draft CRI Guidelines, 2025,

https://www.ipindia.gov.in/writereaddata/Portal/Images/pdf/Draft_CRI_Guidelines_Publication_March2025.pdf

[ii] Telefonktiebolaget Lm Ericsson (Publ) vs Lava International Ltd [CS(COMM)65/2016].

[iii] Biswanath Prasad Radhey Shyam vs Hindustan Metal Industries Ltd (AIR 1982 SC 1444).