Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

ROLE OF ARTIFICIAL INTELLIGENCE IN

INTELLECTUAL PROPERTY RIGHTS

Ananya Singh1, Law College Dehradun, Uttaranchal University, Dehradun, Uttarakhand

Khaleeq Ahmad2, Assistant Professor, Law College Dehradun, Uttaranchal University,

Dehradun, Uttarakhand

ABSTRACT

Artificial Intelligence has profoundly impacted daily life and work, but the
legal system that governs intellectual property rights, such as Copyright and
Patent laws, has been slow to adapt to technological advancements. In this
paper, I discuss the challenges that AI poses to existing IP laws and consider
whether reforms or reinventions are necessary to ensure their continued
effectiveness in the AI era. One issue with AI-generated works is
determining who should own the copyright because it may be challenging to
attribute them to human authors. Additionally, AI systems can produce deep
fakes, which are manipulated images or videos that mimic real images, but
are actually fake, making it difficult to determine the authenticity of such
works. Another challenge for patent law is determining who should own a
patent for AI-generated inventions, as AI systems can invent or improve
existing inventions in ways that surpass human capabilities. Reforming the
IP rules to address AI’s unique properties. My research explores the linking
of patent and copyright rules to AI. This study analyzes these rules as barriers
and starting points for new technologies under IP law. It examines whether
existing laws hinder or align with the purpose of new technology.

Keywords: Artificial Intelligence, Intellectual Property Right & AI,

Copyright Act, Patent Act, Protection of AI System.

1
Final Year Student, Law College Dehradun: Uttaranchal University (Email- [email protected] )
2
Assistant Professor, Law College Dehradun: Uttaranchal University, Dehradun, Uttarakhand (Email-
[email protected] )

Page: 1318
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

I. INTRODUCTION-

In our childhood, most of us have been addicted to certain video games and as a child, I had
always thought that in most of the games, the person playing in front of me was a real person
and managed to beat me all the time. As I grew up I got to know that it wasn't a real person but
a Computer aka Artificial Intelligence playing against me, which got me thinking about how
smart and capable it was and how it is growing drastically in the upcoming times. In this
research paper both of the most growing aspects in their fields Artificial Intelligence in
Technology and Intellectual Property Rights in Law, are clubbed together as Intellectual
Property Rights in Artificial Intelligence3.

The notion of conferring legal personhood upon artificial intelligence (AI) provokes intricate
discussion and debate. Although legal personhood is typically designated to living beings,
corporations, and other organizations that are legally recognized, some have made the
proposition that specific AI systems with advanced capabilities ought to be acknowledged as
legal persons due to technology's rapid progression. The use of an IP system by SMEs in the
pharmaceutical industry largely depends on the company’s business strategy, size, resources,
innovative capacity, competitive context, and field of expertise4. Proponents of granting legal
personhood to AI argue that doing so could lead to greater protection and accountability for AI
systems and their creators. Acknowledging the personhood of artificial intelligence could
potentially trigger the establishment of a regulatory framework for AI decision- making,
availing ethical and responsible behavior from these automatons. Nonetheless, there exist
consequential misgivings regarding the plausibility of endowing legal personhood to Artificial
Intelligence. Within certain intellectual circles, it is believed that the implementation of this
strategy could potentially lead to a skewed allotment of biased favoritism toward synthetic
intelligence mechanisms instead of their flesh-and-blood equivalents. Granting AI systems legal
personhood could have detrimental effects on fundamental rights and privileges enjoyed by
humans as a whole. Moreover, the recognition of AI as a legal entity could have far-reaching
legal and economic ramifications, such as potentially enabling AI systems to possess and
transfer property, enter into contracts and initiate or defend legal cases in courts. While there is

3
Dr. Anusuya Yadav, A complete manual on Intellectual Property Rights & Cyber Law in India 78 (Notion
Press, Chennai, 1st edn., 2022)
4
Rajnish Kumar Singh, Role of intellectual property rights for SMEs: Need to manage knowledge, available at:
http://www.dehradunlawreview.com/wp-content/uploads/2020/06/3-Role-of-intellectual-property-rights-for-
SMEs-Need-to-manage-knowledge.pdf (Visited on March 12, 2024)

Page: 1319
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

no widespread acceptance of AI as a legal entity, recent advancements in this field have

occurred.

In 2017, Saudi Arabia granted citizenship to an AI system called Sophia, which sparked
discussions and controversy regarding AI-possessing personhood. Additionally, certain
countries, such as France and Canada, have recently modified their legal frameworks to
acknowledge specific AI-driven innovations as lawful entities, although these
acknowledgments have limitations in scope.5.

A compelling argument in favor of granting AI legal personhood is that it could promote greater
transparency and accountability in the decision-making processes of AI. For instance, if
artificial intelligence software is recognized as a legal entity, it could be held accountable for
its actions, and any negative consequences could result in legal consequences for its designers.
To address this issue, it is essential to implement responsible and ethical practices for the
development and deployment of AI systems. However, there are also significant concerns
regarding the potential implications of recognizing AI as a legal person. One of the most
pressing concerns is that such an action could result in excessive reliance on AI mechanisms at
the expense of human beings.

II. ABOUT ARTIFICIAL INTELLIGENCE-

What I believe is that Artificial Intelligence (AI) is already here as opposed to some people
believing that it is coming in the future. We humans created machines many years back but just
recently they got smart enough to do most of the work with just programming. Not all machines
are made or work like that. Artificial Intelligence has evolved and will keep evolving in the
coming future.

Artificial Intelligence is a bunch of codes programmed together to make an algorithm for the
exact purpose of software or some particular product. It is believed by Humans that AI
technology is the end of the world, and the particular reason for this belief is that this is what
movies have shown us that way that machines are going to take over the world and we are going
to be their slaves, but it is really funny according to me as for that to happen we humans only
have to program them in such a way that they can have emotions or can mimic emotions, which

5
Purvi Pokhriyal, Amit K. Kashyap, et.al. (eds.), Artificial Intelligence Law and Policy Implication 198
(Eastern Book Company, New Delhi, 1st edn., 2020)

Page: 1320
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

technology has not yet been discovered. AI was imagined or introduced to make the lives of
humans easier by making machines do all the work6.

The world is moving towards the full automation of everything with the least human interaction.
To think about it, the idea of fully automated machines is not a bad one because from what we
have seen the machines are more precise about things and can do work better than us in most
scenarios. As this is a fairly new field there aren't many laws to govern the same, and for that,
we need to get to know more about it.

III. HOW THE INTELLECTUAL PROPERTY RIGHTS APPROACH THE

ARTIFICIAL INTELLIGENCE?

According to the global copyright community, AI systems are based on software, and all the
typical computer program IP issues connected with generating software apply. Today, it goes
without saying that "programs are not merely text; they also behave," and although original
thought and creation are still primarily human abilities, ever-more-powerful computers are
invading. Giving artificial intelligence systems protection, and maybe more problematically,
protecting their creations poses serious problems for the exact purpose of the legal system. The
ascent of artificial intelligence (AI) has unveiled various predicaments concerning intellectual
property rights (IPR), such as apprehensions about the eligibility for patents of AI inventions,
safeguarding copyrights for content generated by AI systems, and preserving trade secrets
related to classified information produced through these machines7. As AI continues to become
more prevalent, the legal frameworks governing IPR are adapting to keep up.

One of the primary ways that IPR approaches AI is through patent law. The laws that regulate
patents apply to AI innovations similarly to any other type of creativity. There's an increasing
number of verdicts from courts concerning the potential for legal protection in patenting this
unique field. To fully realize the benefits of AI technology while minimizing its potential harms,
one must carefully scrutinize and consider its ethical implications.

6
Jasim M, AI for Beginners, A Conscience guide to ChatGPT, Bard, Bing AI & Dall-E 34 (Notion Press,
Chennai, 1st edn., 2023)
7
Rituraj Bhowal, Artificial Intelligence & Law 138 (Central Law of Publication, Allahabad, 1st edn., 2023)

Page: 1321
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

IV. CAN NON-HUMANS BECOME AUTHORS TO PROTECT ARTIFICIAL

INTELLIGENCE CREATIONS?

The subject of whether and how autonomous machine creations should be protected is brought
up by recent developments and significant expenditures made by tech businesses in AI systems.
After years of disappointment, self-learning and deep-learning systems are now surpassing
humans in several fields, largely as a result of the massive growth in the amount of data that
can be fed into the systems8. As a result, AI, for instance, no longer requires engineering inputs
to produce new inventions. The temptation to provide artificial intelligence (AI) systems the
same status in intellectual property law as humans has grown as computer algorithms and
machine learning have emerged as new sources of inventions and creativity. The use of artificial
intelligence in the creation of an invention is irrelevant to its patentability under European
patent law. While under U.S. law "whoever invents or discovers any new and useful process,
machine, manufacture, or composition of matter, or any new and useful improvement thereof,
may obtain a patent," a patent shall be granted when the result is "new, involves an inventive
step, and is susceptible of industrial application." Although the invention's patentability is
unchallenged, it is widely acknowledged that only a natural person may be an inventor under
both EU and U.S. law, and Consequently, businesses, organizations, computers, and all other
non-humans are. This definition excludes because of the inability to possess moral and property
rights.

The main question still is who owns the rights to the creation of an Artificial Intelligence
system? The creator of an artificial intelligence-generated work, such as a piece of music, a
piece of literature, or an animated film, as well as the user, the AI device itself, or a mix of the
three, are all possible candidates. In our analysis, we will take into consideration only works
that have been independently generated by AI, as authorship of works generated by humans
with the aid – through input, guidance, or support – of AI, is recognized definitely9. The issue
of whether a fully automated machine, which by definition cannot be creative, might be given
ownership of a work if it lacks the creativity element, is crucial since the automation of the
creative process falls into a legal grey area. The current copyright office policy is to deny claims
for works that were not created by humans. The purpose of IP law is to give the creator a
financial reward while preventing unauthorized third parties from using the content for

8
Kevin D Ashley, Artificial Intelligence And Legal Analytics 25 (Cambridge University Press, 1st edn., 2017)
9
Anil Shete, A rise of Artificial Intelligence and its Impact on IP Law, available at:
https://www.legalserviceindia.com (Visited on March 15, 2024)

Page: 1322
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

commercial gain. Since AI appears to only be a tool for the exact purpose of creating content,
it does not seem to have any of these requirements and therefore not be protected by intellectual
property laws. According to the same logic, computer software would also need to be granted
IP rights if AI were to be entitled to them. In this case, if an AI program were to translate a
document or correct a spelling error, we might not be able to say with certainty that the computer
has the copyright to the final product. The academic community is divided over the release of
independently generated AI creative works into the public domain, and some, on the other hand,
think that it will limit innovation, the particular reason for the circumstance is that the
businesses that have invested in the development of AI machines won't be able to enjoy
protection or the financial benefits associated with it, eventually discouraging them from
investing. Undoubtedly, this point is as compelling as the first, but it is also true that we are
once again attempting to fit a novel idea into old legal frameworks.

The usage of such old categories would be implied by the claim that the ideal solution must
carefully balance private property rights and the public domain. Despite this, it still seems that
these are the only theoretical ideas that have been significant up to this point. To maximize time,
energy, and financial investments in creative efforts while reducing restrictions on the public's
freedom to enjoy the products of such creativity, policymakers must strike the right balance.
Underestimating this would probably result in problematic levels of either excessive or
inadequate intellectual property protection10.

V. PROTECTING ARTIFICIAL INTELLIGENCE SYSTEMS -

There is still a commonly held belief that patents cannot be used to protect computer software.
Software, like artificial intelligence, varies from computer hardware in that it consists mostly
of a series of instructions, where each code directs the hardware to perform a certain task, and
the computer complies with the rules by producing control signals. Pure software code was not
allowed to be patented in the United States for a long time and the particular reason for this
circumstance is that, as a "mathematical algorithm," it was either regarded as natural law or an
abstract concept. In the 1980s, though, the U.S. concerning people attempting to patent software
ideas, the Supreme Court started to adopt a less rigid theoretical approach.

10
Woodrow Barfield, Research HandBook on the Law of Artificial Intelligence 48 (Edward Elgar Publishing
Ltd, 1st edn., 2018)

Page: 1323
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

The pure mathematical formula or the algorithm itself is not patentable, according to Justice
Rehnquist, but something else. Just because an idea involves an algorithm or mathematical
formula doesn't make it patentable. The Court's reasoning was condensed into the Freeman
Walter-Abele test, which, in essence, was a way to differentiate claims from algorithms
combined with normally statutory subject matter that contain only algorithms. The Federal
Circuit even reduced subject matter patentability as a significant barrier to software
patentability in the 1990s, denying that If not labeled as purely abstract ideas or "useful" truths,
every software creation that produced a useful effect would be subject to a test and made
patentable11. The legal system is currently being forced to reconsider the proper function of
patentable subject matter, even though its breadth has recently reversed and commentators
expect the U.S. The proper limits of the patentable subject matter will shortly be reviewed by
the Supreme Court; a precise line has yet to be determined.

Artificial intelligence, which includes machine and deep learning systems, is merely another
type of software solution that doesn't solely rely on a linear set of programming instructions;
instead, it has an unrecognized ability to "think" for itself. The majority of the software
industry's business is selling its Intellectual Property Rights, as opposed to relying solely on the
protections that actual goods are afforded by property law. For some businesses, this line of
work has evolved into a stand-alone profit center that is frequently independent of the main
services and goods provided by the same software company. Software patents are increasingly
seen as a crucial corporate asset in the expanding software sector; Therefore, the predictability
of its protection under patent law has a major impact on investment choices and, subsequently,
on the long-term viability of a corporation.

VI. IMPORTANCE OF ARTIFICIAL INTELLIGENCE IN INTELLECTUAL

PROPERTY-

As the world evolves and our reliance on technology grows, experts have noted the growing
relevance of artificial intelligence in the IP industry. Many years ago, experts did not
comprehend the importance of excellent data and how it might significantly create and set the
groundwork for the rising IP industry. Professionals in the sector have realized the necessity of
quality data, which eliminates errors, inaccuracies, and inefficient data and cleanly processes

11
Karen Keroli, AI and the Law Developing and deploying AI in Enterprise 138 (O’Reilly Media, 1st edn.,
2021)

Page: 1324
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

the data. The key underlying idea of quality data is quality solutions to IP challenges since a
business demands high- quality data to make all major choices. The use of artificial intelligence
has eliminated the traditional patent search procedures that were previously used to get patents.
The move is more toward AI solutions since it frees up experts' time to focus on analysis,
improved formulation of successful tactics, and insights into patent search. IP regulations
guarantee that it focuses on innovative AI ideas for which patents may be secured in various
nations based on novelty, industrial protection capabilities, and not being too obvious12. With
the introduction of AI, the increased demand for the exact purpose of gaining trademarks,
copyrights, or patents became tough, but blockchain assures the increasing issues encountered
by IP offices. Blockchain is a collection of blocks and chains, each of which holds a piece of
information and allows for verified chains. If one block of information has to be altered, it must
have the approval of all allowed parties. It may aid in a variety of IP operations like record
keeping, management and tracking of distributors and users, evidence of rights, creating
patents, and IP contracts. The main advantage of blockchain in patents is that it may be used to
patent innovation in a variety of nations due to the presence of smart contracts.

VII. INTELLECTUAL PROPERTY RIGHTS LAW IN INDIA-

Artificial Intelligence systems will be impacted by laws like the Copyright Act of 1957 and the
Patents Act of 1970 of India. Aspects of the aforementioned actions are sorted out in this study,
and their effect on AI is reviewed. The development of these AI systems is frequently hampered
by certain restrictions in these laws, which deny Copyright protection for the works that these
devices produce. It is time for these laws to be changed to integrate more advanced and
sophisticated technologies. To foster inventive advancements, while concurrently safeguarding
IPR proprietors from infringement, India has erected a robust and comprehensive legal
framework within which intellectual property rights are protected. The country's system of IPR
is governed by several legislative codes including the Trademarks Act, Copyright Act, and the
Designs Act in addition to the Patents Act among others. In the year 1970, legislation called
The Patents Act was passed to manage and enforce patents within India. It provides for the grant
of patents for inventions that are new, involve an inventive step, and are capable of industrial
application. The Act also outlines the rights and obligations of patent holders and includes
provisions for compulsory licensing in certain circumstances. Under the purview of the

12
Shreya, Vinamra Kothari, Artificial Intelligence in the world of IP, available at: https://iiprd.wordpress.com
(Visited on March 16, 2024)

Page: 1325
 Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

Trademarks Act, in India lies the responsibility for managing and securing trademarks. The act
was initially passed in 1999 to carry out these duties effectively. Inscribed within its provisions
is the accomplishment of trademark registration, provided that said marks are unmistakable and
can serve to differentiate among commodities or amenities offered by one individual in
comparison with those presented by another. The aforementioned statute also accommodates
the revocation of trademarks under specific circumstances, notably if they remain dormant
beyond a prescribed duration. The legislation known as the Copyright Act, initially
implemented in 1957, oversees safeguarding creative expressions such as literature, visual
artwork, and melodies. The system facilitates the enlistment of copyrighted materials and
defines the entitlements and duties associated with copyright possession. Contained within the
provisions of The Act lies a series of measures aimed at safeguarding both performers' and
broadcasters' rights alike, ensuring their lawful protection.

VIII. COPYRIGHT ACT, 1957 AND ARTIFICIAL INTELLIGENCE-

For an expression to be considered original, only one thing must be true: it must not be
plagiarized. Therefore, the author should independently write the piece.

Regarding how to determine whether a work is original, there are two doctrines. As follows:

1. The Sweat of the Brow Doctrine

2. Modicum of Creativity

Section 2(d) of the Copyright Act, of 1957 is a section that puts in doubt the copyright protection
of works created by AI systems. The term "author" is defined in this section. The person must
be an "author" to be the proprietor of any copyrighted work. AI has a challenging problem
because they are typically not viewed as legal persons13. "Author" is defined in Section 2(d) as:
With any literary, dramatic, musical, artistic work or cinematographic works which is computer
generated, the person who causes the work to be created.

In cinematographic works, piracy usually takes place through unauthorized reproduction of the
film in video graphic forms and/or displaying the video through cable networks without the

13
The Copyright Act, 1957 (Act 14 of 1957), s. 2(d)

Page: 1326
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

producer’s permission14. The term "the individual who causes the work to be created" poses a
challenge to this notion. For a human proximity of the creator to the work is crucial to causing
it to be created, and for this, an actual person or a legal person is referred to here. Thus, the
Artificial Intelligence system is not covered by the present Copyright Act. As a result, under
India's copyright laws, the authorship of works produced by Artificial Intelligence would be
ambiguous. Typically, copyright is used to protect creative and literary works. Given that
creation is one of the modern applications of AI the analysis of copyright in the context of
artificial intelligence, becomes relevant. The ambiguity around the position on AI is not new
and extends back while referring to the National Commission on New Uses of Copyrighted
Works in Technology in One of Its a report claiming that the creation of an AI capable of
independent work creation is theoretical and not doable. In 1986, the Office of Technology
Assessment looked at the problem once more when it assessed how IP might be affected by the
rapid improvements in interactive computing. OTA disagreed with CONTU and proposed that
AIs be regarded as legal co-authors of works protected by copyright. When a situation is at its
best, one side will argue the computer's incapacity to be as imaginative as humans, whereas,
on the grounds of defining creativity, some disagree.

Even if nations acknowledged providing copyrights to works created by AI, it is unclear who
will be granted such rights. It is tough to understand, due to the existing state of the legislation
demands, that a right holder be a legal person, something that unless its creator is given that
permission on its behalf, AI lacks. However, there is a flaw in the same, and that flaw relates to
what would occur if the AI system was purchased, and either the inventor or the purchaser will
be given copyright.

IX. PATENTS ACT, 1970, AND ARTIFICIAL INTELLIGENCE-

Patent and the current law:

Section 2 (p) of the Patents Act, 1970 defines the term “patentee”.

"Patentee" means a person for the time being entered on the register as the grantee or proprietor

14
Rajnish Kumar Singh, Indian Approach on Copyright Protection for Cinematograph Films, available at:
http://www.dehradunlawreview.com/wp-content/uploads/2020/12/Paper-3-INDIAN-APPROACH-ON.pdf
(Visited on March 18, 2024)

Page: 1327
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

of the patent. Section 2 (t) defines “person interested”15.

“Person interested” includes a person engaged in, or in promoting, research in the same field as
that to which the invention relates:

Section 6 prescribes the list of persons who can apply for a patent. (a) Any person claiming to
be the true and first inventor of the invention.13 Section 2 (y) of the act defines the term "true
and first inventor". It does not include either the first importer of an invention into India or a
person to whom an invention is first communicated from outside India. The fact that Section
2(y) does not expressly require that the "true and first inventor" be a human opens the door for
the inclusion of works created by AI systems.

However, since the definitions for terms like "patentee," "person interested," and so forth
specify that it must be a person (a legal person), it can be inferred that the legislature intended
the act's overall goal to favor people and other legal persons16. Therefore, these laws must be
changed to meet the needs of an evolving society and scientific systems. If such pertinent
changes are not made to enactments in a very dynamic subject like intellectual property, India
would continue to be a developing nation. These are not unrealistic objectives for a nation like
India.

Patent and Artificial Intelligence:

In today's world, the link between patent laws and AI is growing. As demonstrated in the earlier
section of this paper, AI has been heavily utilized to streamline the execution of fundamental
operations, mainly reducing human effort. At first look, AI-enabled systems appear to operate
similarly to primitive calculators and other similar technology. Though it operates in a far more
complex way. Systems are prepared to carry out operations using their key. Learnings increase
the likelihood that they will develop anything17. Although this is a significant technological
advancement, from a legal perspective, it raises fresh, difficult concerns, From a patent legal
standpoint.There is no doubt that AI will continue to advance with each day that passes.
Businesses like GE, IBM, Apple, and others, increasing their efforts to transform technology

15
The Patent Act, 1970 (Act 39 of 1970), ss. 2(p), 2(t).
16
Kislay Tarun, “Patent of Artificial Intelligence and Its Legal Implications” 2, Brillopedia Journal 8-12 (2022)
17
Mayank Tyagi, Patentability of Artificial Intelligence Creations: Issues and Challenges, available at:
http://www.dehradunlawreview.com/wp-
content/uploads/2020/02/8_Patentability_of_Artificial_Intelligence_Creations-79-87.pdf (Visited on March 18,
2024)

Page: 1328
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

relating to the provision of advanced technology and software solutions based on AI will
inevitably lead to more of these kinds of "inventions18." There is a vast amount of room for
politicians. To create rules for identifying these circumstances, providing it is the most suitable
type of legal protection. With the possibility of using thousands of AI-enabled networks that
operate with or without human intervention, it is necessary to grant some humanoid agent patent
protection so that they can be identified in the event that the invention malfunctions or
potentially violates the law, putting the inventor at risk of criminal prosecution. It must be kept
in mind that in the effort to adapt IP laws to evolving technology, one cannot choose to shift the
balance by minimizing the desirable consequences of criminal laws, which inevitably depend
on the presence of human factors. Additionally, we can't rely on AI because that would diminish
the importance of the human species as a whole.

X. CHALLENGES FACED BY ARTIFICIAL INTELLIGENCE IN INTELLECTUAL

PROPERTY RIGHTS-

A number of potential solutions emerged to address the issue at hand when AI-related ideas
presented a threat to the IP sector in terms of patents, copyrights, and trademarks. Due to AI's
inability to handle massive volumes of data and lack of verifiability in the hands of the
appropriate parties, it did not show to be useful. Many problems with ownership and contract
difficulties are raised by technical issues in AI. In the post, we'll go into further depth on this.

Issues in Contracts: Earlier AI systems mimicked the precise operations of the human brain
using custom hardware, but as software graphic units have been created, there has been a rise
in the usage of central processing units. Therefore, in addition to the aforementioned
components, IP-related problems also surfaced elsewhere when the software was employed.
Challenges faced by business agreements come when there is no language in the contracts
stating the newest evolving software including ownership and license concerns. Agreements
must include all necessary provisions pertaining to third- party authorization, indemnification,
and new IP development software.

Collaborative model: When third parties need pertinent training data to establish themselves
in the market or get pertinent rights to the solution, many AI developers run into problems.

18
Rajnish Kumar Singh, Patenting computer related inventions: India in comparison with US and UK, available
at: http://www.dehradunlawreview.com/wp-content/uploads/2020/06/3-Patenting-compute-related-inventions-
India-in-comparison-with-US-and-UK.pdf (Visited on March 20, 2024)

Page: 1329
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

Having IP rights over cutting-edge solutions guarantees that ownership and licensing will take
on paramount importance, that third parties won't be involved when AI developers pay a license
fee to obtain training data, and that the developers of AI will collaborate with other parties to
safeguard their software interventions.

Customer information: The vendor offers customers who need the assistance of these training
databases to operate in concert with the vendor's software to adapt to customers' business
services necessary training data authorization. When the cyber security system of the customer's
current software given by the seller is breached, issues occur, which frequently raise the issue
of ownership or copyrights. If the clients desire to resell the software to another service
provider, there will once more be a legal issue. It won't be difficult to safeguard software if IP
rights have been gained for it, but if not, it will be challenging for vendors to protect their AI
ideas.

Ownership: As AI has advanced to the point that it can develop 3D designs, print graphics,
writes poetry, and creates artwork, some have questioned the validity of the work generated by
AI and have emphasized the necessity for AI to be protected by IP. It became crucial to
safeguard and protect AI ideas because human inventions are already covered by the system of
IP laws. It was very difficult to be verified under IP due to technical concerns like the software
innovations and concepts used to construct training software. There have been questions raised
about whether AI developers should receive the same compensation for the output stage if they
receive patents for their input stage. Should AI creations be considered in the public domain,
and if so, what criteria will be used to determine authenticity and the scope of the patent's
protection? Their answers most likely stem from the idea that AI advancements improve human
life by offering solutions across the board, but that these advancements need to become accepted
in society as a whole in order for AI creators to be awarded IP protection.

Issues with liability: Just as 3D art, poetry, and other inventions invoke copyright infringement,
there are some AI technologies that analyze business strategies and investment decisions for a
company and use specific AI software that may be patented by the owners but the service
provider would be unaware of the fact and the particular reason for the circumstance is that AI
patents are recognized as being in the public domain, which could bring up the issue of patent
infringement in the AI world. Who is responsible, then, is the issue that arises? In order to
properly respond, it is first necessary to identify the original source from which the copyrighted

Page: 1330
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

technology was copied. Secondly, it is also necessary to determine if the AI's guardians and in-
charge authorities may also be held accountable for patent infringement. It is crucial to prove
the authorities' responsibility so that infringement cases may be handled. Because it is necessary
to establish the aforementioned innovation in the legal sphere, the owner of the invention who
did not secure a patent may also run into difficulty.

Legislation: In order for patented AI inventions to be recognized on a legal basis, it is critical

that IP laws be updated periodically. The IP sector has undergone significant change, and new
inventions and their owners face challenges that require new reforms in order for real owners
to be able to patent or copyright their inventions. There would be no balance between AI
inventions and IP rules if the gap between AI and IP persisted. The need to create forums that
handle AI and IP conflicts on their own is ongoing.

Whether AI robots may be considered inventors is the major issue facing the IP regime. Sophia,
a humanoid created by Hanson Robotics in Hong Kong, was granted its first citizenship in Saudi
Arabia and expressed her desire to pass on her generation during an interview. It, therefore,
begs the issue, can Sophia's offspring be patented? The rigorous regulations of the European
Patent Offices for patents state that an invention can only be given a patent if an extensive
investigation into both its "technical" and "non-technical" qualities results in an "inventive"
step. Even however, there may be difficulties in obtaining patents for such technologies due to
the fact that these criteria vary across numerous nations.

XI. AI INTERSECTING THE DOMAIN OF IP-

There is artificial intelligence all around us. Humans will eventually have to live without
technology in every aspect of their lives. AI is required in every aspect of life, whether it be in
science or athletics. In IP, AI is a useful technology. Many organizations and sectors that depend
on cutting-edge software solutions are adopting it. AI inventions are protected and regulated by
IP. WIPO has made sure that IP laws begin to acknowledge the necessity to implement and
advance AI techniques within every firm. In order to conduct business efficiently, both AI and
IP entail the usage of technologically based ideas that are managed by huge organizations. IP
rules can only serve to safeguard AI developments and legality concerns while also having a
substantial beneficial influence on the economy, the production of goods and services, and their
distribution. AI and IP are related since both deal with innovations that employ patents,
trademarks, and copyrights and because the primary objectives of IP legislation are to foster

Page: 1331
 Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

creativity and discovery in technological fields. Developers of AI software are drawn to the
necessity to patent their innovations, which again overlaps with IP since patents are intellectual
property that must be secured. Since AI is employed in every aspect of society, it is improbable
that it would escape the purview of intellectual property rules. In fact, nowadays every
advancement in the field of AI is covered by an IP clause and offers a practical solution to
pressing problems. The discipline of medicine contains one such significant example of an
intersection. As strategic advancements resulted in a breakthrough in skin cancer treatment and
experts called for the need to reform new technology policies regardless of whether the
technology utilized is patent-eligible or not, the necessity to patent AI robots in the area of
medicine became crucial. To examine the topic eligibility of patents for this, the authorities
established a two test policy. Determining if the claims are directed to be patent-ineligible in
terms of nature, abstract concepts, and natural occurrences was the first step in the evaluation
process. Determining which claims comprise a combination of components that result in a
patent eligible system was the second test. These two tests have made it possible for medical
AI ideas to receive a patent for their work, which will transform the medical system.WIPO, via
its conferences and regulations, has consistently emphasized the importance of AI and how it
should be safeguarded by the IP sector and incorporated into business models to avoid
difficulties such as patent or copyright infringement. Machine learning based on electronic data
is based on information, which is the hub of intellectual property and innovation in a global
digital economy.

XII. CASE LAWS-

● Tay AI and Microsoft (2016): Microsoft's Tay chatbot was deactivated 24 hours after it started
publishing obscene and indecent stuff. Tay was created to interact with and learn from Twitter
users. The incident brought up issues with corporate accountability for the use of AI systems
and the requirement for moral safeguards19.

XIII. CONCLUSION-

In Summary, the current state of AI under intellectual property rights is problematic. Although
acknowledging the work produced by AI is a step in the correct direction, the actual issue lies
in its implementation. In our daily lives, we rely on AI in various ways, such as using Siri, a

19
The Tay AI and Microsoft (2016)

Page: 1332
Indian Journal of Law and Legal Research Volume VI Issue III | ISSN: 2582-8878

limited AI system, to assist with everyday tasks. However, the “Creativity Machine”, an
advanced AI system used by the US military for weapon development, demonstrates superior
logical reasoning and original thought. The emergence of more sophisticated AI systems is
inevitable, and will significantly simplify human existence. Denying AI systems IP protection
based on the argument that they are not considered legal persons or humans is no longer
sufficient. To ensure inclusive growth, it is essential to incorporate all scientific discoveries,
including those involving robots, machines, and AI systems. As we move towards a future
powered by AI, it is crucial to be well prepared for this new era of science, which humans have
chosen.

Page: 1333