An Introspection of ChatGPT in the light of Intellectual Property Rights

PRELUDE

Artificial Intelligence(AI) is a major area that calls for the development of policies in India. The flourishing AI industry and the various AI-centric initiatives of the Government make the matter of policy-making regarding AI a matter of extreme importance. While the present policy processes strive to foster the rapid development of AI for the purpose of securing social well-being and promoting economic growth, India tends to lean towards digital inclusion. However, the risks and deficiencies of data-driven decision-making are a massive obstacle to the development and establishment of AI applications.

Although Artificial Intelligence performs such tasks that demand complex cognitive abilities very easily, it can never be said that it will function flawlessly while performing the task. For Example, AI-driven tools like LSI-R (Level of Service Inventory- Revised) or COMPAS (Correctional Offender Management Profiling for Alternative Sanctions) have been heavily criticized for their racist prediction. Hence, the question arises that if the Artificial Intelligence is negligent and ends up creating a faulty result, who is supposed to take the liability? In such a scenario, determining civil and criminal liability of AI activity is of paramount importance.

In the quest to advance language model technology, ChatGPT was developed by OpenAI. It was after the introduction of GPT-1 (2018), GPT-2 (2019), and GPT-3(2020) that GPT was created to improve user experiences by facilitating interactions with the AI system. ChatGPT is often efficiently implemented in the process of content creation. Further, with the aid of improved understanding and creation of language, it behaves in a very human-like manner.

The development of ChatGPT is a result of OpenAI’s ongoing attempts to advance language model technology. Significant improvements in language creation, contextual comprehension, and the capacity to replicate human-like conversations have been made along the way from the first iterations of GPT to the more sophisticated and complex models like GPT-2, GPT-3, and ultimately ChatGPT. humans may anticipate more advancements and developments in ChatGPT and its applications as AI technology develops, significantly transforming how humans engage with AI systems.8

It is pertinent to note that models like Chat GPT are importantly characterized by their ability to work without being able to take cognizance of the output they are creating. It must also be mentioned in this context that in order to create an outcome of high quality, there lies a need for a good amount of training data, and in order to obtain such data, it may lead to an abuse of copyright-protected information. This calls for an examination of the lawfulness of such utilisation of data and the legal principle of fair use involved therein.

OWNERSHIP OF CHAT GPT-CREATED-CONTENT

The technology of Chat GPT is owned by the creators of Chat GPT OpenAI, which is a San Francisco-based company. OpenAI also owns the intellectual property rights associated with such technology. Speaking about the ownership of specific content created by ChatGPT, the concept is quite complicated in that regard. Unlike the traditional form of writing, the content generated using AI involves the alliance of the author and the algorithm of the AI. While the author makes the specifications of the content, the AI, by way of implementing its algorithm, generates properly shaped content.

In a legal sense, the question “To whom does the ownership belong?” may be answered with the aid of the ruling of the historic judgment of Burrow-Giles Lithographic Co. v. Sarony9. In this case, the U.S. Supreme Court had to face the issue of whether a photograph could be a copyrighted work of a photographer, given that the photograph was created by the camera rather than the photographer behind the camera. While delivering the judgment, the Supreme Court thought that although the machine (camera) is responsible for a considerable amount of the work done, the human input is a necessary element for the creation of the work. The U.S. Supreme Court recognized the photographer to be the author of such a photograph and hence, the exclusive ownership belonged to him.

It may be mentioned in this context that rationale persisted even in cases where the machine was responsible for most of the work, since it was recognised that human input was indispensable for the creation of the work.10

In light of this judgment, it can logically be inferred that the ownership belongs to the person who provides the necessary inputs to produce the literary piece and not the algorithm that shapes the output based on such input. However, such ownership becomes questionable when it involves the violation of the copyright of other works.

THE PRINCIPLE OF INTELLECTUAL PROPERTY RIGHTS

Controlled by the Copyright Act, 1957, the Law of Copyright safeguards original literary, artistic, and musical works. Originality is the most essential element that the copyright law aims to protect. It permits the creator to control the distribution, reproduction, adaptation, and public display of their creations. Further, as per Section 17 of the Copyright Act, 1957, the author of a work, subject to the other provisions of the Act, shall be the first owner of his creation.

In case of a creation under a contract of service it has been provided that “a literary, dramatic or artistic work made by the author in the course of his employment by the proprietor of a newspaper, magazine or similar periodical under a contract of service or apprenticeship, for publication in a newspaper, magazine or similar periodical, the said proprietor shall, in the absence of any agreement to the contrary, be the first owner of the copyright in the work, insofar as the copyright relates to the publication of the work in any newspaper, magazine or similar periodical, or to the reproduction of the work for its being so published, but in all other respects the author shall be the first owner of the copyright in the work”11.

However, the application of copyright law to the contents generated by ChatGPT remains a grey area. ChatGPT has a strong capability, based on the nature of its algorithm, to violate intellectual property rights. It is an important legal threat. During the data training of Chat GPT, a good amount of paper, books, and written data were used, which makes it strongly potential to cause the violation of intellectual property rights. It also increases the possibility of causing breach of contractual obligations when it comes to matters of trade secrets.

In order to prevent any form of controversy, the creator shall take necessary steps to obtain the permit for the use of content protected by the copyright law before processing their input through ChatGPT.

LEGALITY OF IMPLEMENTING CHAT GPT FOR THE PURPOSE OF STORY-WRITING

Implementation of Chat GPT for the purposes encompasses a bundle of intelligent behaviours such as gathering new ideas, comprehending various short stories, analysing various plots, making decisions by using analogy, etc. However, such implementation is strongly challenged by the Intellectual Property Law.

As per Section 2(o) of the Copyright Act, a literary work includes computer programmes, tables and compilations, including “computer databases”. The term “databases” was inserted in the definition of the term “literary work” by virtue of the Copyright (Amendment) Act, 1999. However, the term was not defined.

The term data was first defined under section 2(o) of The Information Technology Act, 2000. The provision lays down: –

“Data means a representation of information, knowledge, facts, concepts or instructions which are being prepared or have been prepared in a formalised manner, and is intended to be processed, is being processed or has been processed in a computer system or computer network, and may be in any form (including computer printouts magnetic or optical storage media, punched cards, punched tapes) or stored internally in the memory of the computer.”

This term has further been defined under section 2(h) of The Digital Personal Data Protection Act, 2023 as “a representation of information, facts, concepts, opinions or instructions in a manner suitable for communication, interpretation or processing by human beings or by automated means.” The term “automated means” signifies a reference to the data created by the implementation of Artificial Intelligence. Hence, it can easily be deciphered that the contents created by Chat GPT fall within the purview of the rights and liabilities of the Copyright Act, 1957.

Also, the term “computer database” has been defined in section 43 of The Information Technology Act, 2000 as: –

“Computer database means a representation of information, knowledge, facts, concepts or instructions in text, image, audio, video that are being prepared or have been prepared in a formalised manner or have been produced by a computer, computer system or computer network and are intended for use in a computer, computer system or computer network.”

Article 10 of the Agreement of Trade-Related Aspects of Intellectual Property Rights (TRIPS) declares computer programmes and compilations of data to be eligible for protection under copyright law as: –

“1. Computer programs, whether in source or object code, shall be protected as literary works under the Berne Convention (1971).

2. Compilations of data or other material, whether in machine-readable or other form, which by reason of the selection or arrangement of their contents constitute intellectual creations shall be protected as such. Such protection, which shall not extend to the data or material itself, shall be without prejudice to any copyright subsisting in the data or material itself.”

The Universal Declaration of Human Rights (UDHR) highlights that intellectual property rights are the same as human rights. In this context, Article 27(2) lays down that “Everyone has the right to the protection of the moral and material interests resulting from any scientific, literary or artistic production of which he is the author.” Further in this regard, Article 15(1)(c) of the International Covenant on Economic, Social, and Cultural Rights, 1966, recognizes the right of everyone “to benefit from the protection of the moral and material interests resulting from any scientific, literary or artistic production of which he is the author.”

In this context, it would be pertinent to observe that whatever data Chat GPT trains for the purpose of getting an output shall conform to the rules laid down by these laws. The protection of the intellectual property rights extends to these databases, and any form of diversion from the rules laid down by these laws would cause a violation of the original author’s economic and moral rights in respect of the ownership of the author’s copyright in literary work. It also raises another important concern that if a book is written with the aid of ChatGPT, the amount of ownership of the person who provides the data by which the AI is trained to produce such a result remains in the dark.

The Delhi High Court, in the case of Tata Sky Limited v. National Internet Exchange of India12, laid down that artificial intelligence may be employed for creative purposes, subject to the restrictions laid down by the laws and the courts relating to infringement and violations.

The Supreme Court, in Justice K.S. Puttaswamy v. Union of India13, recommended that the Union Government examine and put forward a robust regime for data protection.

In August 2023, the Digital Personal Data Protection Act, 2023 was enacted with a view to process digital personal data in a manner that provides recognition not only to individual rights of data protection but also to the need to process such data for lawful purposes. Besides imposing responsibility upon data fiduciaries, the Act mandates prioritisation of Data Principal rights by way of implementing proper organisational and technical safeguards.

PARTING NOTES

While the legal landscape will keep evolving, the issue related to copyright and ownership of content generated by the implementation of artificial intelligence will remain complex. The liability and responsibility of the input and output of Chat GPT will be a point of grave importance due to the privacy and protection of such data imposed by the laws. Maintaining transparency regarding the process and availing license agreements would perhaps be a solution.

Editor’s Note

MeitY / NITI Aayog AI regulation drafts (2024–25)

• India will adopt a principle-based AI governance framework (safety, fairness, transparency, accountability, human-centric).
• AI systems may be classified by risk levels (low, medium, high), with stricter obligations for high-risk uses.
• Developers and deployers must ensure algorithmic accountability through audits, documentation, and reporting.
• Transparency measures such as disclosures about AI use and explainability of outputs are recommended.
• Watermarking / labeling of AI-generated or synthetic content (e.g., deepfakes) should be mandatory.
• A national AI incident database is proposed to track harms, failures, or misuse of AI systems.
• Data protection and privacy must align with the Digital Personal Data Protection Act, 2023.
• MeitY may establish a Technical Secretariat and inter-ministerial body to oversee AI regulation.
• Sector-specific rules (health, finance, education, etc.) will complement the general framework.
• Drafts emphasize innovation-friendly regulation, including sandboxes and voluntary compliance, but binding AI law (AI Bill) is under consideration.

Tags: English Section, Kalpabiswa, Ritojit Dasgupta, দশম বর্ষ প্রথম সংখ্যা