What Are the Ownership Rights When AI Creates Inventions – Can AI Be Listed as an Inventor?

The present research article seeks to explore various facets of the relationship that does exist between artificial intelligence & its implications upon the subject matter of intellectual property rights all across the globe and notions with respect to the same being shared by various courts across the countries.

IPR

Shashank Upadhyay

10/28/20254 min read

INTRODUCTION

The introduction of Artificial Intelligence has revolutionized the very landscape of creativity & innovation. From the point of drafting legal documents & formulating music to design the pharmaceuticals & developing Algorithms, AI systems have demonstrated a very remarkable ability to perform tasks that once required Human ingenuity. This automatically raised the very question of fundamental importance in the subject matter of IPR worldwide i.e. “when AI Creates an invention, who owns it and can A.I be categorised as an Inventor?

The issue before us is not merely academic in nature. It sits at the very intersection of technology, law & ethics. Challenging the very foundation of patent jurisprudence, which traditionally presumes humans as inventors.

LEGAL DEFINITION OF INVENTOR

Under the national & International Patent systems, the very concept of “inventor” is inherently human. Let’s say for instance,

1. The Indian Patents Act, 1970: which actually does not explicitly define the term ‘inventor’; however, its language implies human involvement. S.2(1)(y) defines a ‘patentee’ as the person for the time being entered on the register as the grantee or proprietor of the patent. The entire procedural structure of patent filing i.e. is submission, declaration & signing presupposes involvement of human agency.

2. Similarly, the same is there in case of ‘United States Patent Act’ (35 U.S.C. §100(f)) which goes on to define inventor as “the individual or, if a joint invention, the individuals who invented or discovered the subject matter of the invention.” The terminology “individual” is interpreted strictly to be construed as a natural person.

3. The European Patent Convention (EPC) follows an approach based upon comparison which requires that the inventor must be a natural person.

This traditional interpretation excludes machines or non-human entities, thus placing AI outside the very purview of the legal definition of ‘inventor.’ However, this conventional stance has already been tested followed by recent developments. Most notably, the DABUS cases.

THE DABUS CONTROVERSY: A.I AS AN INVENTOR

A very crucial case on this aspect is that of the DABUS controversy, which speaks for ‘Device for the Autonomous Bootstrapping of Unified Sentience,’ an AI system developed by Dr. Stephen Thaler. The DABUS autonomously formulated two generated inventions. The food container & a flashing light device. The patent was filed in the name of A.I. as an inventor.

Consequently, the same filing exaggerated the global debates, where the courts in the U.K., U.S., E.U., Germany & Australia subsequently rejected the very notion of A.I. inventorship on the ground that the law even requires a human inventor.

In the case of Thaler v. Comptroller-General of Patents, 2021, it was held that “only a person can be an inventor” under the Patents Act, 1977, as delivered by the U.K. Court of Appeal.

In the case of Thaler v. Vidal, 2021 delivered by the U.S District Court for Eastern District of Virginia held that, ‘an inventor must be a Natural Person’.

The European Patent Office also reiterated the very fact that only the humans can have the legal personality & consequently, the rights & obligations associated with the inventorship.

However, not in all the jurisdictions took a uniform stance as in 2022, The Federal Court of Australia briefly recognized AI Inventorship before the decision was overturned on appeal. Similarly, South Africa became the first country to grant a patent listing DABUS as an inventor, though its Patent Examination System is largely formal & does not inculcate Substantive review in itself.

CONCEPTUAL & JURISPRUDENTIAL ISSUES INVOLVED IN THE PRESENT ISSUE

The very reluctance to recognize Artificial Intelligence as an Inventor involves various jurisprudential & Conceptual concerns.

1. LEGAL PERSONALITY: AI System lacks legal personality as they cannot own property neither enter into the contract nor can sue or be sued. Since the patent rights are proprietary in nature, it follows that only a legal entity is capable of owning the property (a person) can hold them. listing AI as an inventor raised the unsolved question i.e. “who would hold the patent rights on behalf of an A.I”.

2. ATTRIBUTION & ACCOUNTIBILITY: Inventorship entails within itself not just credit but, the very accountability & responsibility of it, a human inventor can be held accountable for fraudulent claims, ethical violations, or misrepresentation. A.I, by the very contrast, cannot be held liable or accountable under the very legal system.

3. HUMAN CREATIVITY & MORAL RIGHTS: Patent law historically views invention as an act of human ingenuity which is the manifestation of human creativity & intellectual labour which is nothing but, recognizing A.I as an inventor who challenges the philosophical foundation of intellectual property, which is based upon human intellect.

POLICIES & ETHICAL CONSIDERATIONS

This debate is not merely confined to the legal domain but is also policy driven. Identifying A.I. as an Inventor could have profound implications, such as

1. Encouragement of Innovation: which actually allows A.I.-generated inventions to be patented, which could stimulate technological progress by providing the incentives to develop more advanced A.I. systems.

2. Risk of Monopolization: On the other hand, unrestricted patenting of AI-generated inventions could lead to the concentration of IP rights in the hands of a few corporations controlling advanced AI systems.

3. Erosion of Human Creativity: Overemphasizing A.I. inventorship may undermine the societal value placed on human creativity and intellectual effort.

4. Ethical Dilemmas: the question which arises under the present sub – portion is that whether A.I. does really possess intent, originality or creative authorship including qualities which are traditionally associated with human consciousness.

PROPOSED SOLUTIONS & WAY FORWARD

In order to address the evolving challenges, several legal reforms & frameworks have been introduced to tackle with the same;

1. Amending Patent Laws: The legislature could redefine “inventor” to include human-A.I. collaborations, where A.I. is being acknowledged as a contributing entity rather than being called the legal inventor.

2. Providing a New Categorization of I.P Protection: Govt.’s are invited to introduce Sui generis protection for A.I. Generated Inventions, similar to date base rights found in European Union (E.U).

3. Requirement of Disclosure: Patent Applicants could also be required to disclose the extent of A.I. involvement in the inventive process, ensuring transparency without altering the human-centric model.

4. Attributing rights of AI owners: ownership could also be automatically vested in the person who owns or facilitates the operation of the AI system, making them responsible for the AI invention.

5. International Harmonization: The scenario of innovation has already been showcased, which is harmonizing the AI inventorship policies through WIPO or TRIPS, which could prevent such jurisdictional inconsistencies present in the system.

CONCLUSION

The question of whether Artificial Intelligence can be listed as an inventor or not. Exposes fundamental tension between technological advancement & the legal framework available designed for a human-centric world. At present, the consensus across major jurisdictions remains clear: Artificial Intelligence cannot be recognized as an inventor because it lacks legal personality & moral accountability. However, the law must evolve to accommodate the practicalities of AI-driven innovations.