While India’s IP system has made significant strides in digital accessibility, Rushil Verma, in his submission for the SpicyIP–Jhana Blogpost Writing Competition 2025, explains how it continues to fall short on intelligibility. He suggests that integrating “Explainable IP” could help transform patent and trademark databases from mere document repositories into inclusive, user-centric public knowledge systems. Rushil is a fourth-year B.A. LL.B. (Hons.) student at Symbiosis Law School, Pune, with a focused academic and professional interest in Intellectual Property Law.
Beyond PDFs and Portals: Building ‘Explainable IP’ for a Transparent and Inclusive Indian Innovation Ecosystem
By Rushil Verma
Introduction: Access Without Understanding
India’s IP system has undergone an impressive digital transformation. Over the last decade, the Office of the Controller General of Patents, Designs and Trade Marks (CGPDTM) has moved steadily toward an online-first environment, where patent filings, legal status records, and the prosecution documents are available through the Indian Patent Advanced Search System (InPASS). This transition from manual archives to searchable online databases has been previously discussed on SpicyIP, including in Prof. Shamnad Basheer’s examination of the need for more intelligible open access to Indian Patent Office records.
Yet, for many IP practitioners, this progress conceals a deeper problem. India has succeeded in making a significant amount of IP information available than was the case previously, but it is not yet understandable. Inventors, MSMEs, university researchers, law students, journalists, and public-interest groups still face enormous difficulty interpreting what this information actually means. The documents are there, but the meaning is locked within legal and technical language, unstructured metadata, inconsistent prosecution histories, and opaque claim amendments.
Therefore, it can be said that India has indeed succeeded in making digital portals, however, what it still lacks is explainability.
Why Explainability Matters: Moving From Data to Knowledge
The idea of explainability first gained traction in AI governance, where regulators emphasised the need to make algorithmic decisions understandable to the public (see this piece by Auste Simkute et al). The same logic applies to India’s patent and trademark information systems.
In simple, concrete terms, explainable IP refers to patent and trademark data organised in a way that enables non-experts to understand not just what an invention is, but what its claims mean, why examiners objected, how those objections were resolved, and what scope the final protection actually covers.
Currently, an inventor attempting to navigate InPASS to examine or search for patent application(s) encounters a maze of documents, including the First Examination Report (FER), amended claims, technical diagrams, multiple forms, applicants’ and opponents’ submissions, inconsistent metadata, and prosecution documents that require specialist literacy. The EU-India IP Helpdesk has also highlighted practical challenges faced by applicants when using Indian patent databases, noting that searching Indian patent data may require significantly greater effort and that platforms such as InPASS lack bulk-download functionality necessary for further data analysis.
Explainability, therefore, represents a shift in philosophy: from merely hosting documents to communicating them.
India’s Digitised IP Landscape: Progress With Persistent Gaps
Notwithstanding these issues, India has made significant strides. The IP prosecution documents that were once accessible only through physical file inspections can now be viewed online. IP scholars like Abhishek Kumar et. al. in their work “Developing Metadata Standards for Efficient Searching of Indian Patent Information: A Plan” have also discussed the need for metadata improvements and standardisation, observing that inconsistent bibliographic data affects retrieval quality and analysis.
However, these three systemic shortcomings continue to impede meaningful access to IP information:
- First, information friction persists. A typical FER is dense, lengthy, and unintuitive to interpret. While part of this difficulty arises from the inherently technical nature of patent examination, variations in drafting style and lengthy narrative objections can further complicate comprehension for non-specialist users attempting to understand the reasoning behind objections or the logic of amendments.
- Second, metadata fragmentation undermines searchability. Inventor names may appear under multiple spellings or formats across records, for instance, institutional applicants may be listed as both “Indian Institute of Technology Delhi” and “IIT Delhi”. Similarly, corporate assignees may appear under abbreviated and expanded forms, such as “Reliance Industries Ltd”. and “Reliance Industries Limited”, preventing reliable portfolio tracking. Patent records may also continue to reflect outdated or inconsistently updated International Patent Classification (IPC) codes, a common situation where applications remain tagged under older subclasses despite later revisions, making technology-specific searches and sectoral trend analysis more difficult (See this piece by Stephen Kunin). Without clean data, IP analysis becomes guesswork.
- Finally, the prosecution context is almost invisible. Users see a stack of documents, not a story. Why were certain objections raised? How did the amendments resolve them? What prior art mattered? What claim scope survived? Without context, online access becomes little more than an administrative display.
Accordingly, integrating explainability into the digitalised IP system is essential to ensure that IP information is not merely available, but intelligible and functionally usable.
The Case for Explainable IP: Democratizing Innovation at Scale
India’s innovation ecosystem is expanding rapidly. With over 63 million MSMEs, thousands of technology startups, and growing research output from universities, the demand for accessible IP information is increasing, but the capacity to interpret that information has not kept pace.
For instance, a small manufacturing entrepreneur in Coimbatore exploring whether her mechanism infringes an existing patent cannot meaningfully understand the scope of prior claims without expensive legal help. A student researcher in Patna studying renewable-energy technologies may know how to retrieve documents, but not how to interpret examiner objections or claim dependencies. A journalist investigating pharmaceutical patents may struggle to track how claim amendments circumvented Section 3(d) of the Patents Act, 1970.
Explainable IP empowers these users. Plain-language summaries of claims would help innovators determine whether an invention is new or potentially infringing. Structured prosecution histories would allow civil society and journalists to trace how an application evolved. Clean metadata and classification consistency would enable researchers to conduct meaningful trend analysis.
Explainability transforms IP databases from specialised repositories into public knowledge systems.
Global Lessons: India Has an Opportunity to Lead
While no major IP office has yet fully implemented explainability, several global initiatives point in that direction. The IP5’s Global Dossier integrates prosecution histories across the US, Europe, Japan, Korea, and China, enabling cross-office comparison in a single interface Google Patents adds further layers by providing cited references, prior-art exploration, and semantic search, hinting at how visualised, structured IP data can enhance public understanding.
Academic research also demonstrates what is possible. The “TechNet” patent-semantic network project shows how natural language processing can map relationships between inventions, fields, and prior art, dramatically improving interpretability. India could leapfrog global systems by making explainability a deliberate design principle, not just an incremental improvement.
Challenges and Realistic Expectations
Explainability is not without challenges. Some of the assumptions behind it deserve nuance.
One assumption is that improving metadata will automatically democratize innovation. In reality, legal literacy and resource constraints still matter. Explainability must be paired with outreach and education, particularly for MSMEs and public institutions. Another concern is administrative burden. Requiring examiners to produce plain-language summaries manually could slow examination. Solutions must therefore be automated, drawing on natural language generation tools or standardised templates.
There is also a risk of oversimplification. Patent claims carry legal precision; reducing them to plain language must be done carefully to avoid misinterpretation. But this is a design problem, not a conceptual flaw. Addressing these concerns ensures that explainability strengthens, rather than distorts, the integrity of India’s IP system.
A Practical Blueprint for Explainable IP
A realistic roadmap for explainability does not require rebuilding the IP office. It requires adding a new interpretive layer to the data already available. This could begin with standardising metadata, issuing unique inventor and assignee identifiers, and harmonising classification codes with updated IPC standards to ensure consistency across patent records. Clean data is foundational for all downstream explainability efforts.
Next, prosecution histories can be visualised as timelines that reflect examiner objections, applicant responses, amendments, and legal status changes, allowing lay users to grasp the procedural journey at a glance. Plain-language abstract summaries of inventions and claim scope, generated through automated text-simplification tools, could be made available in English and major Indian languages. In addition, audio narration of such summaries, enabled through contemporary speech-to-text and text-to-speech systems, could significantly improve accessibility for users who may not be comfortable navigating dense written records. This mirrors India’s broader digital public infrastructure philosophy, serving the “next billion” users through inclusion rather than complexity.
Finally, periodic machine-readable data releases would allow researchers, journalists, civil-society groups, and civic-tech developers to build independent explainability tools, dashboards, and analytical models, much like the open-data initiatives developed through India’s Open Government Data Platform (OGD). Explainability does not replace legal expertise. It complements it by widening the base of users who can meaningfully participate in innovation.
Conclusion: India’s Moment to Build a More Understandable IP Future
India’s digital transformation is well underway, but information alone does not guarantee empowerment. True transparency requires clarity, not just availability. A patent system that only experts can interpret is a patent system that excludes the majority of potential innovators.
Explainable IP offers a path toward an innovation ecosystem where knowledge is shared, participation is broad, and understanding is not limited by technical literacy or geography. It represents a shift from “documents stored online” to “knowledge accessible to all”.
As India aspires to lead globally in science, technology, and digital public infrastructure, building explainability into its IP systems is not merely a technical improvement, it is a democratic one. A system that explains itself is a system that invites participation, scrutiny, collaboration, and creativity.
In moving toward an explainable IP framework, India has the opportunity to create the most transparent, inclusive, and innovation-friendly IP ecosystem in the world, one where understanding is not a privilege, but a public right.