Abstract
Amidst the Fourth Industrial Revolution, India’s emergence as a significant hub for artificial intelligence (AI) innovation, marked by a dramatic surge in AI-related patent filings, confronts an intellectual property (IP) framework ill-equipped for a new technological paradigm.
This article critically examines the doctrinal and practical incongruities between India’s extant IP laws and the realities of AI-generated innovation, hypothesizing that the current legal regime’s insufficiency creates a grey area that obstructs technological advancement and investment.
Through a methodological approach combining doctrinal legal analysis, comparative case law review, and policy analysis, the study maps critical legal gaps and operational challenges.
Key findings reveal significant impediments, including statutory bars to patenting algorithms and ambiguous authorship provisions for computer-generated works within India’s patent and copyright laws.
Analysis of recent landmark judicial and administrative decisions concerning AI inventorship and the patentability of computer-related inventions underscores the judiciary’s struggle to apply analog-era laws to autonomous creations.
These legal uncertainties are compounded by systemic challenges within the Indian Patent Office, including application backlogs and a deficit in specialized technical expertise.
The paper’s principal contribution is the articulation of a multi-pronged reform agenda, recommending targeted legislative amendments to clarify AI inventorship and authorship, advocating for enhanced IPO capacity, and proposing consideration of a sui generis protection regime.
It further suggests implementing algorithmic impact statements and a national registry for AI works to foster transparency.
Ultimately, this research argues that proactive, technology-neutral legislative reform, supported by empirical investigation into innovation trends and stakeholder needs, is imperative to secure India’s position as a global leader in responsible AI development.
Finally, this study registers that active and technology-agnostic legislative modernization with the empirical support of research into trends and demands of the industries and stakeholders is necessary to reinforce India as one of the global pioneers in responsible development of AI techs.
Introduction
The Fourth Industrial Revolution is transforming the economy and society, with Artificial Intelligence (AI) reshaping various aspects of life. AI is influencing human creativity and ingenuity, leading to complex legal and ethical issues, particularly in Intellectual Property (IP) rights. Traditional frameworks are being challenged, and ownership of AI output is disputed.
This research critically analyses the relationship between AI and IP in India, focusing on its fast-growing technology industry and potential to become a global innovation hub. Overcoming AI challenges and opportunities will significantly impact India’s economic growth and technological competitiveness.
This discussion explores AI-driven innovations and challenges in protecting intellectual property. It examines the Indian Patent Office’s (IPO) issues with AI-generated applications and the impact on India’s innovation system. The article evaluates the legal framework, case law, and policy discussions, suggesting a future approach without compromising India’s IP system values.
India’s emergence as a significant player in the global AI patent landscape reflects both the country’s technological prowess and the pressing need for legal clarity in this domain. With AI-related patent filings growing exponentially—from approximately 5,000 applications during 2010–2015 to over 90,000 applications in fiscal year 2024—India has secured its position as the fifth-largest patent filing nation globally.
Definitions and Concepts
Artificial Intelligence (AI) – A broad field of computer science that deals with the creation of intelligent agents, which are systems that can reason, learn, and act autonomously.
AI encompasses a wide range of technologies, including:
- Machine Learning (ML): A subset of AI where algorithms are trained on large datasets to identify patterns and make predictions without being explicitly programmed.
- Deep Learning (DL): A more advanced form of ML that uses deep neural networks to learn from vast amounts of data. It is the technology behind breakthroughs in image recognition and natural language processing.
- Generative AI: A class of AI algorithms that can generate new content, such as text, images, music, and code. Models like GPT-3 and DALL-E have gained attention for producing human-like creative outputs.
Practical Challenges in Protecting AI-Generated Innovations
Authorship and Inventorship:
A fundamental requirement for obtaining both copyright and patent protection is the presence of a human author or inventor. However, when an AI system independently generates a work or an invention, the question of who should be recognised as the author or inventor becomes highly contentious.
Several possibilities have been debated:
- The AI itself: Some argue that if an AI is truly autonomous, it should be named as the inventor. However, current legal systems, including India’s, do not recognise non-human entities as inventors.
- The developer of the AI: The person or team that created the AI could be considered the inventor.
- The user of the AI: The person who operates the AI and provides the initial inputs could also be a candidate for inventorship.
Novelty and Non-Obviousness
To be patentable, an invention must be new (novel) and not obvious to a person skilled in the relevant field of technology. Assessing the novelty and non-obviousness of an AI-generated invention can be challenging. Since AI can analyse vast datasets, it becomes difficult to determine whether its output is genuinely novel or merely a recombination of existing knowledge.
The standard of a “person skilled in the art” may also need reconsideration in an age where AI can perform tasks once exclusive to human experts.
The Black Box Problem
Many advanced AI systems, particularly deep learning models, operate as “black boxes”. This means it can be difficult, if not impossible, to understand the exact process by which the AI arrived at a particular output.
This lack of transparency poses challenges for patent law, which often requires a detailed disclosure of how an invention works so it can be replicated by others. If the inner workings of an AI’s creative process cannot be explained, meeting disclosure requirements for a patent becomes problematic.
Scope of Application of AI in Innovation
Creative Domains / Works
Generative AI has advanced to the point of breaking the boundaries of the traditional conceptualizations in the creative arts.
Contemporary systems:
- Compose music: Produce music by way of history both stylistic tendencies and generation of new music.
- Create visual art: Turn creative generative networks such as DALL-E and Midjourney to generate visual art out of the outcomes of short written descriptions, generating gorgeous images.
- Write literature and poetry: Write literature and poetry and produce stories and poetry that are indistinguishable to human-made works Collectively, these trends make traditional ideas of attribution, authorship, and copyright more complicated.
Scientific Research and Development
It is the role of Science and Technology to operate AI is speeding up discovery in a wide range of fields in the natural and social sciences.
- Drug development: algorithmic applications could be used to search a large collection of biological data to propose novel therapies, thereby reducing the time line of clinical trial history.
- Materials science: reliably predictable properties of materials can be modelled using algorithms and can be applied to design the next generation products.
How AI Generates Innovations
The Role of Data and Algorithms
At its core, AI-driven innovation is powered by two key components: data and algorithms.
Data: In the area of artificial intelligence, it is a well-known fact that the performance of machine-learning models is data size and data heterogeneity dependent. The quality and diversity of such corpus will then decide to what extent the AI can g in the field of artificial intelligence , it is necessary to note that algorithms refer to a set of procedural rules and statistical models by which the system can learn on its own using the data and then generate new content.
These algorithms can be coded to detect trends, predict, or be creative by combining the existing factors. The reliance on the training data, furthermore, puts into the foreground the major issues of the copyright.
It is also possible to describe the spectrum of human involvement in the AI-based innovation:
- Human creation with the help of AI – where a human uses AI as a creative extension. As an example, an artist may use an AI-based image-editing service to perfect a composition, and thus the artist will be the author of the final image.
- AI creation directed by a human – in this case, the human provides an initial idea or parameter set to the system, and the AI produces the work based on that. This is the case with the “prompt engineering” that forms the basis of generative systems like DALL-E, and the associated authorship question is a controversial one.
- Autonomous AI generation – the most difficult one, where an AI system creates something based on no direct human input or oversight. The most paradigmatic one is the DABUS case in which an AI has made a patent application as an inventor. enervate new and aesthetically interesting output.
Literature Review
Academic Books and Foundational Texts
Although the grand entrance by AI and IP books is decidedly internationally oriented, the intellectual values that they teach do relate to the Indian landscape.
- Ryan Abbott, Research Handbook on Intellectual Property and Artificial Intelligence (2022) – this is what was to be resorted to in the matter. It discusses the entire globe, although various chapters also provide structures that are particular to India. To illustrate, the patentability requirements examined in the book such as novelty and inventive step can be used to deconstruct Indian Patent Act and how it positions inventions by AI. It also suggests a concept of AI legal neutrality, the logic that as long as a person or an AI is engaging in a given behaviour the law shouldn’t treat them differently and that concept will provide us with one very important way of thinking about the problem Indian law will need to address.
- Reasonable Robot: Artificial Intelligence and the Law (2020) another book by Ryan Abbott: This one makes the argument that when AI acts, legally it should not be treated differently than a human acting. The conflict with the natural person rule, which is ingrained in India IP laws, occurs with that idea; such conflict emerges in court judgements and government policies, all the time.
- Artificial Intelligence, Intellectual Property, Cyber Risk and Robotics (2024) edited by Ruth Taplin: an interdisciplinary book (i.e., combination of economics, law and ethics). In the case of India, an already rapid rising nation that aspires to be an AI hub, avoids the threats to jobs and data privacy.
All these concerns are reflected in the policies set by NITI Aayog that demand a balanced system to be established that encourages innovation and maintains these risks at bay.
Legal and Regulatory Framework in India
The core of the Indian debate is rooted in its existing IP legislation, which was enacted long before the advent of sophisticated AI.
The Patents Act, 1970
It is here that the inherent issue in the protection of inventions based on AI is manifested in India. Section 3(k): According to this rule a mathematical or a business method or a computer programme per se or algorithms cannot be patented. Since AI and machine learning involve algorithms, most AI inventions fall against this impediment.
The Indian patent office (IPO) introduced Guidelines of Computers-Related Inventions (CRI), e.g., the 2016 of 2017 to resolve any confusion. According to the guidelines, there are only computer programs that can be patented through their direct association with a unique hardware element and demonstrating a tech related effect or solving a tech related problem. These demands challenge creators of AI solutions not only to demonstrate their concepts in the form of algorithms, but as concrete technical solutions as well. This effectively means that legal experts often criticize these guidelines as being inconsistent, not having the force of law, which leaves the inventor at a loss as to what actually is allowed.
The Copyright Act, 1957
Another set of challenges can be challenged by this law which has its unclear provision. Section 2(d)(vi): This sub part defines an author in the case of a so-called computer-generated work to be the person who causes the work to be created. This fact has brought a lot of controversy under the law. According to some scholars, the provision is foresighted and it can be read in such a way that it gives authorship to a user or developer of an artificial intelligence system. Some say that computer-generated in 1957 was likely referring only to simple, automatic output and not what we get in the state of the art of generative AI today. It is this uncertainty which is at the heart of recent Indian case law.
Requirement of Human Authorship: Despite Section 2(d)(vi), the overarching principle of copyright law is that it protects original works stemming from human creativity. The lack of a clear definition of “originality” for AI-generated works remains a major gap in the literature and the law.
Policy Initiatives by NITI Aayog
The paper did not suggest particular legal amendments even though it recognized the problematic issues of IP. In 2021 NITI Aayog released “Principles of Responsible AI” and among the set it emphasized safety, equality, inclusivity, transparency, accountability in regard to the development and deployment of AI.
Although all these principles are not legally binding, they frame ethical debates concerning AI and tend to be mentioned in legal articles regarding the future regulation of AI operations in India.
Landmark Case Law and Precedents
Recent cases have brought the theoretical challenges into sharp, practical focus.
The DABUS Case (Indian Perspective)
Dr. Stephen Thaler filed a patent application in India naming his AI system, DABUS, as the sole inventor. In 2021, the Indian Patent Office refused the application, stating that under the Patents Act, 1970, only a “natural person” can be recognized as an inventor. The IPO’s decision reinforced the human-centric nature of Indian patent law. Legal commentators have extensively analysed this decision, contrasting it with the varied outcomes in other jurisdictions like South Africa (which granted the patent) and the UK/US (which did not). The case firmly establishes the current legal barrier to non-human inventorship in India.
Ankit Sahni v. Union of India (The RAGHAV Case)
This is the most significant Indian case on AI and copyright. Ankit Sahni, a lawyer and artist, filed for copyright protection for an artwork titled “Suryast,” generated by his AI system, RAGHAV. In 2020, the Indian Copyright Office initially registered the work with Sahni listed as a co-author alongside the AI. This was hailed as a progressive step. However, the registration was later withdrawn for re-examination following public debate. This case highlights the profound uncertainty within the Indian Copyright Office on how to apply Section 2(d)(vi) and whether an AI can possess authorship rights. It remains a key case study in all recent Indian literature on AI and copyright.
Ferid Allani v. Union of India
While not an AI case, this decision by the Delhi High Court was crucial for software patents. The court ruled that if a computer program demonstrates a “technical effect” or “technical contribution,” it should not be barred by Section 3(k). This precedent is now vital for applicants of AI-related patents, as it provides a legal pathway to argue that their invention is more than just an algorithm “per se.”
Scholarly Articles and Legal Blogs
A vibrant discussion is taking place in Indian legal journals and online forums.
SpicyIP
SpicyIP: This leading Indian IP law blog provides continuous, in-depth analysis of developments. Its posts on the DABUS and RAGHAV cases, critiques of the CRI Guidelines, and updates on government policy offer real-time insights into the evolving landscape.
Journal of Intellectual Property Rights (JIPR) and other academic journals
Journal of Intellectual Property Rights (JIPR) and other academic journals: These publications feature scholarly articles that delve into the doctrinal and theoretical challenges. Common themes include proposals for sui generis (unique) protection for AI-generated works, comparative analyses of Indian law with international frameworks, and discussions on the economic implications for India’s “Make in India” and “Digital India” initiatives.
Upcoming Legislative Developments
The present signs are that such a sequence of significant legislative milestones has begun, with the first one playing out in 2023, with the release of the Draft Digital India Act. This new tool is a better version of the Information Technology Act 2000, which seeks to offer a more thorough and stronger regulation that would be capable of addressing the contemporary digital challenges.
The second milestone will be in February 2025, when NITI Aayog is expected to publish an in-depth report on AI governance. The document will outline specifications of risk assessment, regulatory compliance, and sector-specific use-cases, taking the policy discussion far beyond the most high-profile commercial deployments of the technology. Besides shaping future trends in the legislative sphere, the report will also determine new industry best practices. At the same time, the Government of India is thinking of establishing an AI Safety Institute. The outlines of a comprehensive AI law are expected in 2-3 years by industry observers, and at the same time NITI Aayog is likely to publish its own guidelines and open them to public consultations.
Challenges of the Patent Office and Implications for Innovation in India
The effectiveness and efficiency with which the IPO handles the patent applications related to AI will have a conclusive impact on the Indian innovation ecosystem at large. Over a number of years, the IPO has been grappling with a huge backlog of unprocessed patent applications. Importantly, the evaluation of the patent applications in the field of AI requires a very high degree of technical knowledge.
To add to these issues is the Section 3(k) of the Patents Act. This non-patentability, excluding the patentability of algorithms and computer programs, in their per se form, contributes to the ambiguities that already existed with regard to AI-related inventions.
Potential applicants should thus be extra careful when going through Section 3(k) by proving the technical impact or contribution that their inventions will have. Whether it succeeds in adjudicating AI-related patent applications will be a deciding factor in the formation of the Indian innovation ecosystem. Such delays and this uncertainty can act as a deterrence to innovation and research and development (R&D) investment.
Limitations and the Future of AI-Generated Innovations
Modern AI is not yet capable of claiming creativity or reflexive consciousness despite the unprecedented hype around it. Pattern recognition and the use of structural recombination are good, but not the small-scale appreciation, purposefulness and emotional complexity that typify human creativity. The next generation systems will be capable of attaining greater autonomy and displaying truly new, even unexplainable, results.
This development will further increase the need of our legal and ethical frameworks to keep up with the change. AI will not substitute human creativity as much as it will be an effective tool that will complement it. The urgent challenge, therefore, facing the society is to channel the positive outcomes of AI to the general population and at the same time establish strong legal and ethical protections to counter the associated dangers.
Conclusion
On the one hand, the AI can introduce new possibilities that have never been seen before in terms of solving some of the most burning issues of humanity; on the other, this technological innovation poses fundamental concerns regarding the existing systems of intellectual property protection. The current study has, therefore, determined that the intersection of AI and IP is a complex area of study with significant consequences to the Indian ambitions of becoming an innovation-led economy.
The existing system of intellectual property, which was created, it should be kept in mind, in a pre-AI world, is simply not ready to follow the fast-paced technological advancements. The uncertainty of whether AI-generated works and inventions are legal or not also creates some level of uncertainty that might not be desirable by the innovators and investors.
These challenges are further highlighted by the issues that the Indian Patent Office is facing, namely, backlogs, lack of experienced examiners and the complexity of the process of interpreting existing legislation.
India must be proactive and visionary in order to realize the best opportunity of AI-led innovation. Legal Reforms: The Patents Act and the Copyright Act must be reviewed and amended in detail, so that there is no ambiguity, as far as the protection of AI generated innovations is concerned.
The possible approaches are the introduction of specific provisions on AI inventorship or authorship or the use of technology-neutral framework that can be used to accommodate future changes. Strengthening the IPO: It is also important to be able to develop capacity of the Indian Patent Office. These measures will not only make the AI-generated innovations more secure, but they will also allow India to reiterate its intention to possess a healthy and fair innovation ecosystem.
Suggestions
Scope of the Problem
AI-related inventions are being filed more frequently, but innovators often face uncertainty over what qualifies for protection and how long it will take for their applications to be examined. Many feel the system is not yet equipped to deal with the complexity of AI. Examiners also struggle to apply current computer-related invention guidelines consistently, and attorneys point out that the rules are not always clear to clients.
Beyond Patents
Not every AI innovation fits neatly into the patent box. Some companies prefer to keep algorithms and data secret. Others are worried about competitors copying AI-driven business processes, which can raise issues of fairness and competition. This shows the need for a more flexible toolkit that goes beyond patents, blending trade secret protection with fair competition principles.
A Sui Generis Framework for AI
India could consider introducing a dedicated AI Innovation Protection framework. This would allow recognition of AI-created works and inventions in ways that traditional patent and copyright systems may not cover. Key features could include:
- Eligibility: Recognition of inventions developed jointly by humans and AI, with disclosure of each role.
- Rights: More limited than traditional patents, but still providing exclusivity for a set period.
- Timeframe: Faster decisions to match the pace of technological change.
- Registration: A simple and transparent registry, separate from standard patents and copyright.
Borrowing from international examples, India might classify AI applications by risk — for instance, stricter oversight for healthcare AI but lighter requirements for chatbots. A requirement for innovators to submit an “Algorithmic Impact Statement” could also be introduced, ensuring disclosure of data sources, steps taken to reduce bias, and methods used to validate the system.
Ethical and Social Dimensions
AI innovation may reduce certain jobs, particularly in manufacturing and services. This makes retraining programs important, so affected workers can shift into roles that are created by AI rather than displaced by it. There is also the ethical need for transparency in how AI systems are developed, especially around the datasets used. Independent audits and strong data governance standards would help build trust.
Data and Competition
India’s data protection framework will affect how AI systems are trained, particularly regarding the use of localized datasets and user consent. Startups may feel the burden of compliance more heavily, while large companies with vast resources may consolidate control through patent acquisitions. It will be important for competition authorities to monitor whether a handful of players gain too much power through control of AI patents and data.
Strengthening Policy and Capacity
The Patent Office could be better supported by hiring experts in AI and equipping them with AI-based search tools to reduce delays. A stakeholder task force under NITI Aayog could also help by drafting amendments to existing laws, ensuring India keeps pace with global debates on AI.
Real-World Learning
Indian startups working in areas like drug discovery or smart manufacturing are already facing hurdles in protecting their innovations. Looking abroad, models like the US system of monitoring AI-enabled medical devices after they enter the market could be adapted, ensuring Indian innovations are both safe and globally competitive.
Future Roadmap
India could also encourage the registration of AI-generated works in a dedicated registry, marking clearly what role humans and machines played. This would help clarify issues of authorship and ownership. Two broad approaches can guide policy:
- Innovation Oversight, where AI outputs are monitored for their risks and impacts.
- AI Legal Neutrality, where AI outputs are treated just like human inventions without special treatment.
References
- World Intellectual Property Organization. “Artificial Intelligence and Intellectual Property.” WIPO. June 2023.
- Thaler, Stephen. “Device for the Autonomous Bootstrapping of Unified Sentience (DABUS) Patent Filings.” Patent Applications, various jurisdictions, 2018–2021.
- National Institution for Transforming India (NITI Aayog). “National Strategy for Artificial Intelligence (Discussion Paper).” NITI Aayog, Government of India, June 2018.
- NITI Aayog. “Principles for Responsible AI.” NITI Aayog, Government of India, September 2021.
- Ministry of Electronics and Information Technology (MeitY), Government of India. “Computer-Related Inventions Guidelines, 2017.” Gazette of India, January 2017.
- The Patent Office, Government of India. The Patents Act, 1970 (as amended). Government of India, 1970.
- The Copyright Office, Government of India. The Copyright Act, 1957 (as amended). Government of India, 1957.
- Ahlawat Associates. “Navigating Intellectual Property Rights in AI Innovation: From Idea to Patent in India.” Ahlawat Associates Blog, May 2024.
- IEEE Xplore. “Intellectual Property Protection in AI-Driven Innovations,” IEEE, 2023.
- Saikia, Pratik. “AI’s Role in Shaping Intellectual Property: Emerging Trends and Legal Implications.” Aequivic, May 2025.
- Zatapult. “Indian Patent Office: Roles, Processes, and Challenges.” Zatapult Insights, 2024.
- International Journal of Law and Technology. “The Impact of Artificial Intelligence on Intellectual Property Rights: A Case for Reform in Indian Patent Law.” IJLT, November 2024.
- National Law University, Nagpur. “AI and Intellectual Property in the Digital Age.” CI–Dec 2024 Publication, NLUNagpur.
- Record of Law. “Artificial Intelligence and Intellectual Property Rights in India: Legal Challenges and Opportunities.” Record of Law, February 2025.
- India Briefing. “Regulation of AI and Large Language Models in India.” India Briefing, March 2024.
- Abbott, Ryan. Research Handbook on Intellectual Property and Artificial Intelligence. Edward Elgar Publishing, 2022.
- Abbott, Ryan. The Reasonable Robot: Artificial Intelligence and the Law. Cambridge University Press, 2020.
- Government of India. Patents Act, 1970.
- Government of India. Copyright Act, 1957.
- Controller General of Patents, Designs and Trademarks. Computer-Related Inventions Guidelines, 2017.
- NITI Aayog. National Strategy for Artificial Intelligence. NITI Aayog, Government of India, June 2018.
