The Impact of Artificial Intelligence on Patents

Abstract

Artificial intelligence is expected to aid the global patent system in various different ways. This will be a natural progression from the current system in which the classification and search of prior art represents an essential stage of patent examination. Throughout this process, successful examination relies on the skill of the examiner to identify highly relevant prior art for a particular application. However, prior art documents can be difficult to find and the volume of information available can be overwhelming. This is especially true in technical fields where patents are often written in highly specific, jargon-laden language. In these circumstances, an AI system that could quickly and accurately identify and classify relevant prior art would be an invaluable tool for examiners. This article will explore the areas in which artificial intelligence (AI) will impact the global patent system. These areas include the utilization of AI as a tool to process patent data in order to aid patent examination, the proposition that AI might become a creator of patentable inventions, and that AI may also impact how patents are enforced. AI is an area of computer science that looks to create systems that can perform tasks that would normally require human intelligence. This encompasses a diverse range of abilities including visual perception, speech recognition, decision making, and translation between languages. AI systems can take various forms ranging from simple tools to expert systems designed to emulate human decision making.

Full text

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 63
Paper Publications
The Impact of Artificial Intelligence on Patents
Dr. Dema Matrouk Aloun
DOI: https://doi.org/10.5281/zenodo.11001028
Published Date: 20-April-2024
Abstract: Artificial intelligence is expected to aid the global patent system in various different ways. This will be a
natural progression from the current system in which the classification and search of prior art represents an essential
stage of patent examination. Throughout this process, successful examination relies on the skill of the examiner to
identify highly relevant prior art for a particular application. However, prior art documents can be difficult to find
and the volume of information available can be overwhelming. This is especially true in technical fields where patents
are often written in highly specific, jargon-laden language. In these circumstances, an AI system that could quickly
and accurately identify and classify relevant prior art would be an invaluable tool for examiners.
This article will explore the areas in which artificial intelligence (AI) will impact the global patent system. These
areas include the utilization of AI as a tool to process patent data in order to aid patent examination, the proposition
that AI might become a creator of patentable inventions, and that AI may also impact how patents are enforced. AI
is an area of computer science that looks to create systems that can perform tasks that would normally require
human intelligence. This encompasses a diverse range of abilities including visual perception, speech recognition,
decision making, and translation between languages. AI systems can take various forms ranging from simple tools
to expert systems designed to emulate human decision making.
Keywords: Artificial intelligence, AI system, global patent system.
1. INTRODUCTION
1.1. Definition of Artificial Intelligence
The core of the whole issue lies in the definition of artificial intelligence. As it stands right now, both the US and the EU
have laws that state a computer program is not an inventor as it is not a natural person, leaving the true inventor as the owner
of the computer program. However, this law spans over a time period where invention by a computer program was not a
reality. When comparing this previous law to the current state of AI, machine learning, and automated systems, we run into
a blur where certain AI can be recognized to have its own decision-making and creative processes separate from the owner,
while others are tools for inventors to use. Since the AI can be considered an automated entity with the owner being a user,
it can also be said that under a work-for-hire agreement, the AI does not retain ownership of its creation. A work-for-hire
agreement occurs when the AI is a tool to fulfill the function of the worker. In the case of US law, the agreement may be
explicit, but under EU law, the agreement is implicit, assuming that the employer will retain ownership of anything created
during employment. If the AI is considered to be a worker, the results of its creations are still owned by someone else. This
would make it so that an inventor seeking to patent AI would not need to patent AI itself but rather the results of its AI and
what the AI had done. However, this would not be different from present laws that state an employee invention is always
the property of the employer and thus does not solve the issue of patenting AI. A slight difference occurs when the AI is
independent, as it would require its own patent for what it is to do. However, since this is a relatively new issue with no
precedent and the changing laws between the US and EU states, any assumptions made are inconclusive.
The "The Impact of Artificial Intelligence on Patents" discusses the relation between AI and patents and argues that the
current patent laws in the US and the EU do not have a proper definition for AI, which has become very advanced and
complex. The author raises concern that if proper steps are not taken now to alter patent laws, it might not only hurt the AI

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 64
Paper Publications
industry but also take a step back in humanitarian advancement by AI in fields like healthcare. The essay is of particular
interest in the current time, as AI continues to become an integral part of human life, from the use of automation in industries
to the personal use of virtual assistants in the form of smartphones.
1.2. Importance of Patents in the Technological Landscape
A patent is a set of exclusive rights granted by a state to an inventor or their assignee for a limited period of time in exchange
for a public disclosure of an invention. Patents provide a temporary monopoly to the patent holder to exploit their invention
for commercial gain, by preventing others from commercially exploiting the patented invention in return for the public
disclosure to stimulate further innovation. Patents have been a part of English law since the 16th century. The system is
designed to encourage technological progress by granting the inventor a form of public recognition and monetary incentive
in return for the time and effort put into researching and developing an invention. The recognition of patents exists in many
forms of technology, from pharmaceuticals to computer software, and are considered essential to the competitive advantage
of companies in a technological market. In essence, patents are a driving force for the innovation of technology to serve
different purposes whether it is purely to gain financial benefits or to help the greater good of society. However, in enabling
patent holders to prevent others from exploiting the patented invention, this monopoly has the potential to hinder
competition and follow-on innovation resulting in a negative impact in many areas of emerging technologies.
2. THE ROLE OF ARTIFICIAL INTELLIGENCE IN PATENT CREATION
An automated intelligent system can be used to search and analyze the existing body of patent art to determine the
patentability of an invention. One method of searching the patent database is to use keywords to find patents related to the
invention in question. Often these simple keyword searches are insufficient because two patents addressing a similar issue
may use different terminology. Intelligent automated systems employing natural language processing or other AI techniques
are able to analyze the content and context of patents in order to better understand the relationship between patents on a
given subject. Additionally, AI systems such as data mining tools or genetic algorithms can be used to analyze large portions
of the patent database, or even the entire database, to find patterns or relationships between existing patents. This
information could be useful in determining areas of high or low patentability, as well as the relationship between a
technology and the social/economic effects of its patent protection.
From the earliest days of the Patent Office, tools for technology have been employed in the process of examining and issuing
patents. The first examinations of patent applications were conducted with the aid of the "Classification of American
Patents," which was a system originally designed for searching and analyzing the existing body of patent art. In the modern
era, patent examiners are burdened with reviewing large numbers of complex patent applications with limited resources and
tools. Automated intelligent systems are being developed to assist examiners in searching, analyzing, and ultimately
understanding the state of the art. These intelligent systems are not only powerful tools for assisting human examiners, but
they are also potentially powerful creators of new patentable inventions.
2.1. Automated Patent Search and Prior Art Analysis
Automated search and retrieval of patent documents have been routine for over two decades. All the same, it has been
estimated that only 50% of the world's patent documents have been translated into searchable text. A still smaller percentage
of these documents are in a readily accessible format. Increasing the accessibility of global patent data is of critical
importance. LSTM recurrent neural networks have been used to translate patent claims into the text of Boolean search
queries. More recently, scientists at the WIPO have been exploring the use of new AI language processing techniques to
summarize and translate patent documents. Automating the process of patent search and prior art analysis is an area where
AI is predicted to make a significant impact. The high stakes and potential cost savings have already attracted the attention
of several start-ups and university research groups. In the US, it is estimated that the recent America Invents Act will lead
to a 30% increase in the demand for prior art searching, due to the new patent validity review procedures. Although it is a
highly skilled task, much of the work in identifying relevant patents and categorizing information can in principle be broken
down into a series of defined steps. This makes it possible to develop software to emulate this process, albeit the current
generation of AI technology requires that the steps be tightly constrained and the data highly formatted. As natural language
processing and machine learning techniques continue to improve, it is expected that the automation of prior art analysis will
become more comprehensive and sophisticated. This will shift the role of human analysts towards quality control and
decision making, much as computerized chess programs have surpassed the playing abilities of all but a few grand masters.

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 65
Paper Publications
2.2. AI-assisted Patent Drafting
2.2.1 The drafting of a patent application is a complex task and the patent specification - a comprehensive description of
the invention - is a critical part of the process. Patent applications are generally drafted by patent attorneys or patent agents,
often with the participation of the inventor. Preparing a quality patent specification can be a time-consuming and expensive
process; a reduction in cost would make it more attractive for individuals or small businesses to seek patent protection. AI
programs have been developed to analyze the content of a technical document, and to classify the content as belonging to
one of a number of predetermined categories. The IBM patented technique is an example of such a program. While such
programs can be useful as drafting tools, without a human understanding of the invention in the first instance and the
categories that the program seeks to classify, there is a danger that the resultant classification and generated claim could be
wide of the mark. Further, the predefined nature of the categories in such programs limits their usefulness in dealing with
innovative technology for which there is no existing terminology. Machine learning AI has the potential to assist greatly in
reducing the cost of patent drafting. Using statistical techniques to enable a computer to learn from a large set of example
documents, machine learning AI can automatically generate predictions or decisions. This might be achieved by training
the machine on a set of existing patent specifications and teaching the machine to predict the classification of sentences or
paragraphs of text in the specification. If successful, such predictions would allow automation of the process of categorizing
an abstract, or specific sections of the specification, to be covered by intelligentsearching404 on later. Alternatively, the
machine could be trained to produce descriptive text classified in a certain way, for example a detailed description of an
apparatus and a set of instructions for making and using the apparatus. Although still in its infancy, AI generated text holds
great promise for the future.
2.3. Intellectual Property Protection for AI-generated Inventions
Since many AI techniques, particularly those that lead to automation in tasks traditionally performed by humans, continue
to advance, it is more than likely that there will be a growing number of AI-generated inventions. This being the case, a
solicitation from the AI community for less restrictive IP laws will ensue. In view of the widely known role that patents
have in the economy to incentivize innovation, it is sensible that there is some level of incentive for inventors to use AI
technology and subsequently patent the inventions. This would involve all types of invention whether AI-generated, or use
of AI as a tool to assist in the inventive process. An incentive system could involve lessening the fees in the patent
application process for AI inventions, or giving extended patent-terms for AI inventions in particular fields. Any changes
to the system must be determined by comparing the costs and benefits to society and the economy, and must be done so
over a long term with longitudinal study of its effects.
According to the ordinary rule for the purpose of vulnerability of a patent that is the publication under the consent, it is
conceivable that AI automation could create an understanding of an invention, which could be fit to an invention being
derived. The inventiveness of AI inventions will be contrasted with human-created inventions and central concepts of both
patent and copyright law. It could be recommended that some kind of enterprise where potential AI inventions are disclosed
for the benefit of making a catalog of them to decrease the hindrance of the search for prior art. If AI inventions are totally
exempt from protection, society overall could possibly be deprived of future innovations that could be notably beneficial.
Therefore, the policy making and law for IP protection of AI inventions needs to be cautiously thought out and balanced.
3. CHALLENGES AND CONSIDERATIONS IN AI-RELATED PATENT APPLICATIONS
Inventorship and ownership of AI-generated inventions In terms of traditional patents on chemical, electrical, and
mechanical technologies, a standard patent application requires the naming of the inventor or joint inventors. The
application must also include an oath or declaration stating that the inventor believes he/she is the first to invent the claimed
invention. Under US law, an inventor is defined as an individual who has conceived of the invention. To satisfy the
requirements for inventorship, it is crucial that a human inventor has provided a significant contribution to the conception,
design, and reduction to practice of the invention. Inventorship cannot be ascribed to a person that merely contributed an
idea of a desired result, general outline, or some collocation of known elements, from which the invention had not been
specifically designed. This issue concerns the patenting of AI-generated inventions as they are computers or machines, not
humans, which are developing and producing new and innovative ideas. If an AI program was solely responsible for creating
an inventive product or process, it is arguable that the computer would be considered the inventor. United Kingdom law
defines an inventor as the actual deviser of an invention, thus an AI program could be regarded as an inventor if presented

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 66
Paper Publications
with the legal definition. This has been demonstrated in the recent application by Dr. Stephen Thaler (founder of Imagination
Engines Inc.), who listed his Artificial Inventor Creativity Machine (DABUS) as the inventor in patent applications filed
with the US Patent and Trademark Office and the European Patent Office. When applying for a patent, it is the norm for
invention rights to belong to the employer of the inventor or the person who has commissioned the work leading to the
invention. This will also raise issues of assignment of AI-generated inventions as machines become more prevalent tools in
research and development. The owner of the rights would be the person considered to be the inventor and is therefore tasked
with proving inventorship when applying for a patent. This again could lead to disputes as to who or what is considered to
be the inventor and whether the rights would be ascribed to it.
3.1. Inventorship and Ownership of AI-generated Inventions
In the AI context, identifying the "inventor" is not straightforward. It is generally agreed that the inventor is the entity that
contributed to the conception of the inventive ideas or reduction to practice of the patented invention, which is in line with
current patent law. However, in cases where the inventive concept was initially a set of broad instructions as to how to solve
a particular problem and the AI autonomously adapts this concept such that it is now inventive, there is no clear-cut answer
on whether it is the programmer, the AI, or a combination thereof who is the inventor of the AI-improved solution. This is
a crucial question both for incentivizing AI innovation and for clarity on patent ownership. In most cases, it is likely that
the AI will be deemed to be a tool used by the programmer and the inventive concept that the AI came up with as an
improvement to an existing idea. This is because it is the programmer who will have directed the AI to solve the problem
in the first place. The reasoning that the AI is a tool may mean that the act of using AI is implicitly separating from AI
research and development, and in turn may mean that there is less innovation in AI itself and more use of AI to solve specific
problems. This is due to the fact it will be less economically viable for research into an AI when the improved ideas are
owned by the user rather than the developer. For example, a medical researcher funds AI to find a cure for a disease, but
the results are owned by the drug company said researcher is contracted to.
3.2. Patentability Criteria for AI Inventions
Hence in the different areas of technology, the examiner needs to have an understanding of what might constitute common
general knowledge, the prevailing prejudices and preferences in the field, and the "state of the art" from which the inventors
in the field are making incremental advances. In some areas such as computer implemented inventions, the examiner may
need to have an understanding of not just what is the current technology, but where the technology is heading and what
problems are being addressed. This means there is a significant burden on the examiner and the EPO is always looking for
ways to make its search strategy and procedures more efficient and effective, so that an adequate search can be carried out
in a reasonable time.
Formulation of the inventive step may be such that the claim is expressed as a method or process. In other cases, particularly
in the field of computer science, the claim may be to a system, for example a "search engine", a "data retrieval system", or
to a "computer program product". In other cases, especially in mechanical engineering or in the field of medical devices, it
may be to a product such as a kitchen utensil or a stent, or even to a new use of a known substance, product or process. This
means that there is no single search strategy that will identify all the potentially relevant prior art.
3.3. Ethical and Legal Implications of AI Patents
An interesting situation is present in the case of universities which employ professors and students to do research, and all
inventions by them are owned by the university. The situation where the AI is to replace the human tool comes under the
case law of the State Street Bank wherein it is held that anything that makes an act must be a human, and if AI is to be a
substitute for a human in the near future, then inventions by AI will be granted patents.
As a researcher, it is early to decide as to whether AI as machines having intelligence can be an inventor or not, but the
instances in which AI may be considered as an inventor exist in various fields. In the US, various patent statutes define an
inventor to be any individual who contributes to the conception of the invention. This is further explained by many
judgments, and AI being a tool can be conceived to be inventing a few things. Even if the AI is not considered as an inventor,
the owner of AI will be creating a vicarious liability wherein the AI will be treated as an employee doing acts on instruction,
and in this case, the act done by the employee to implement the employer's instructions is all attributed to the employer. So
the owner of AI will be responsible for infringement by AI, and any patents from inventions by AI will be owned by him.

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 67
Paper Publications
Ethical and legal implications of AI in the context of patents can be understood and addressed in different aspects. These
imply to the context as to how AI as an inventor can be made liable for infringement of right and how the AI can itself have
rights on the patented invention. The other aspect comes into picture is about the subject matter for which the inventor seeks
a patent. In this context as to whether the low or the already developed AI may be able to patent the invention, the second
instance may be whether AI as a tool in the hands of an inventor has contributed enough to make the invention patentable,
or whether the AI has been used to implement the invention.
3.4. International Perspectives on AI-related Patents
The international nature of AI technology and patenting thereon presents unique challenges that have to date not been fully
addressed. Some of the most important questions on international harmonisation of AI patents were raised in an open letter
to the current European Commission on the draft proposal of a statutory instrument to the effect that it shall be permissible
to grant patents for computer implemented inventions. The letter, authored by leading academics in the field raises the
question of whether or not patentability criteria for inventions developed through AI should be different to those applied to
inventions by a natural person, suggesting that in the long run there is a risk of 'biasing innovation in favour of AI on the
assumption that it will always be an improvement', if the same criteria are used. This argument has not yet been fully fleshed
out, especially considering that areas in which patentability criteria is not met could be important areas where development
of AI technology would be beneficial to mankind and it is not clear that there is any benefit to be gained from preventing
AI development in the context of the public at large. This raises the further question of costs versus benefits to society,
should it transpire that the costs of patenting AI technology are too high relative to an assumed few marginal benefits, many
companies and inventors may be dissuaded from further AI development, or may simply forgo patenting with knowledge
of near certain ineligibility. An inference can also be drawn that the general laws on patentability are insufficient for AI
technology and the fear of loss of international harmonisation has lead to speculation of different AI patent systems, such
as the proposal that there should be a formal IP rights subsystem within the Berne Convention for AI created works. An
increased development in AI patents may effect changes in costs upon international patent applications, agreements between
countries, already seen with the patent cooperation treaty might need revision with AI specific provisions. With even the
farthest reaching effects of the WTO TRIPS agreement not yet fully understood, policy makers have much to deliberate
before fundamental changes to the existing international patent system.
4. FUTURE OUTLOOK AND RECOMMENDATIONS
When considering the future of AI and its patenting potential, the need for flexibility within patent laws and institutions is
crucial. It is inevitable that the next decade will bring rapid change to AI, as software and algorithms will continue to take
on a more prominent role in innovation and discovery. Machine learning, as mentioned earlier, is already having a
significant impact on the nature of innovation itself. Developments in this field, such as those driven by deep learning, will
likely shape the technology of the future. These innovations will be highly unpredictable, even to their creators, and as the
capabilities of AI surpass human abilities in an increasing number of fields, there will be growing pressure to automate the
inventive process. An example to consider is the potential for full AI authorship of patentable inventions. For the present,
this is prohibited in many countries, as current laws require an inventor to be a 'natural person'. However, it is not difficult
to imagine that AI authorship may one day be economically significant, and there will be a need to reassess the fundamental
concept of inventor. Another important issue to be revisited in the future is that of patentable subject matter. This will be
especially relevant in the context of public policy considerations surrounding the promotion of innovation and the public
interest. As AI becomes an increasingly prevalent tool for simulating human cognitive processes, there will be a growing
trend in the development of AI methods to replace human activities in various fields. The general consensus among legal
scholars is that current patent laws are ill-equipped to handle inventions produced by purely computational processes, and
studies have shown that the patentability of software and business methods varies greatly between jurisdictions. At the
international level, there is an ongoing debate regarding the TRIPS agreement, which mandates minimum standards of
patent protection for WTO member states. In spite of this, the flexibility of patentable subject matter is quite large, and the
precise boundaries of what may be patented are determined through national laws and legal precedent.
4.1. The Evolution of AI and Its Impact on Patent Systems
The AI and machine learning panorama has existed and evolved over decades. Though the sphere of patenting for AI
innovations is comparatively new, the implementation of patent law to AI-based innovations is an evolution of a process

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 68
Paper Publications
that has concerned the very nature of AI. Because the development of AI technologies has been driven traditionally by the
tutorial and analysis community, this sort of software system has been traditionally set within the sphere of public domain.
The goal has been to make a platform for collaboration and a typical infrastructure that can be improved and changed by
any range of researchers. A good example is the standard technique that the analysis community treats software system
developments. Usually, protecting software implementations isn't seen as one of the necessities of the development method.
An engineer or programmer could develop a piece of code as proof of a concept for a new technique, but this research code
isn't considered the innovation itself. Usually, these researchers are successfully implementing their new concepts into
existing software environments. In other cases, it may take years of more research and development before the approach
becomes commonplace. Usually, it's only when the community is mostly united on a specific technique that some sort of
software implementation is considered the technique itself and not a failed experiment.
4.2. Policy and Regulatory Frameworks for AI Patents
Without the appropriate frameworks in place, it is possible that the patent system will fail to incentivize AI-based innovation
and output. Should the inventor of an AI-created invention not be considered the 'owner' of the invention, there may be little
motivation to disclose said invention to the public: without public disclosure, there can be no patent. An alternative scenario
exists where it is companies rather than inventors who are considered the owners of AI creations. This could lead to a new
system of internal R&D patents where only large firms have the resources to patent AI inventions.
However, the global nature of AI research and development makes it difficult to predict the direction of AI and the nature
of AI-created inventions: international policy has therefore been slow to react. Achieving AI patent harmonization across
borders will be an arduous process. In order to best shape future policy, it is essential that legal scholars work alongside
scientists and engineers in the AI field.
The AI technology has evolved significantly over the last few decades. With the rise of ML, AI has the capacity to "invent"
new solutions to complex problems, and with AI systems increasingly used across numerous industries, there will be a
corresponding increase in AI-based innovations. Current global patent systems, however, were not designed with AI in
mind. With AI-created IP stretching the boundaries of inventorship and machine autonomy challenging concepts of
infringement, current patent systems may fail to accommodate AI-based innovation. As such, it is likely that there will be
substantial reform to the patent systems of the world to best facilitate AI innovations.
4.3. Collaboration and Knowledge Sharing in AI Patent Space
To foster collaboration and knowledge sharing within the AI patent ecosystem, open, freely accessible patent databases are
critical. Unlike common patents, AI patents usually involve heavy utilization of data and machine learning models, making
them difficult to be disclosed completely in the traditional way. Trade secret protection might not be appropriate for these
AI patents when data and models involved need to be shared with the public for verification and reproducibility. This is
especially important for AI in healthcare, where patient well-being is at stake. Forcing these patents to be disclosed without
the right environment might lead to premature disclosure or sub-optimal application, creating a negative impact on patentees
and the public. It is important to establish international norms that encourage these AI patents to be disclosed in the form
that's best for knowledge sharing and provide flexible options of protection to patentees. As demonstrated by the Open
COVID Pledge (OCP), where stakeholders pledge not to assert patent rights for activities aiming to end the pandemic, a
collaboration-friendly environment can be provided by collective actions of patentees and users. Norms and options should
cater to diverse needs and concerns arising from different technologies and industries AI is involved in.
REFERENCES
[1] R Setchi, I Spasić, J Morgan, C Harrison… - World Patent Information, 2021 - Elsevier. Artificial intelligence for
patent prior art searching. cardiff.ac.uk Cited by 23
[2] AV Giczy, NA Pairolero, AA Toole - The Journal of Technology Transfer, 2022 - Springer. Identifying artificial
intelligence (AI) invention: A novel AI patent dataset. [HTML] Cited by 51
[3] L Jiang, S Goetz - arXiv preprint arXiv:2403.04105, 2024 - arxiv.org. Artificial Intelligence Exploring the Patent
Field. [PDF]
[4] S Baruffaldi, B van Beuzekom, H Dernis, D Harhoff… - 2020 - oecd-ilibrary.org. Identifying and measuring
developments in artificial intelligence: Making the impossible possible. researchgate.net Cited by 104

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 69
Paper Publications
[5] DM Kang, CC Lee, S Lee, W Lee - Proceedings of the 24th Symposium …, 2020 - dl.acm.org. Patent prior art search
using deep learning language model. [HTML] Cited by 14
[6] WK Robinson - Nev. LJ, 2020 - HeinOnline. Artificial intelligence and access to the patent system. unlv.edu
Cited by 12
[7] K Vaish, P Rawat, S Kathuria, R Singh… - … Intelligence and …, 2023 - ieeexplore.ieee.org. Artificial Intelligence
Reducing the Intricacies of Patent Prior Art Search. researchgate.net Cited by 1
[8] S Yanisky-Ravid, R Jin - Social Science Research Network, 2020 - ncbi.nlm.nih.gov. Summoning a new artificial
intelligence patent model: in the age of pandemic. nih.gov Cited by 15
[9] SS Tu - Cardozo Arts & Ent. LJ, 2020 - HeinOnline. Patenting Fast and Slow: Examiner and Applicant Use of Prior
Art. wvu.edu Cited by 15
[10] J Lee, S Park, J Lee - IEEE Access, 2022 - ieeexplore.ieee.org. A fast and scalable algorithm for prior art
search. ieee.org Cited by 2
[11] A Engel - GRUR International, 2020 - academic.oup.com. Can a patent be granted for an AI-generated
invention?. [HTML] Cited by 14
[12] ME Leusin, J Günther, B Jindra, MG Moehrle - World Patent Information, 2020 - Elsevier. Patenting patterns in
Artificial Intelligence: Identifying national and international breeding grounds. sciencedirect.com Cited by 14
[13] JJ Bryson, H Malikova - Global Perspectives, 2021 - online.ucpress.edu. Is there an AI cold war?. delorscentre.eu
Cited by 16
[14] J Drexl, R Hilty, L Desaunettes-Barbero… - Max Planck Institute …, 2021 - papers.ssrn.com. Artificial intelligence
and intellectual property law-Position statement of the max planck institute for innovation and competition of 9 april
2021 on the current debate. mpg.de Cited by 15
[15] D Kim - GRUR International, 2020 - academic.oup.com. 'AI-Generated Inventions': Time to Get the Record
Straight?. oup.com Cited by 47
[16] S Baruffaldi, B van Beuzekom, H Dernis, D Harhoff… - 2020 - oecd-ilibrary.org. Identifying and measuring
developments in artificial intelligence: Making the impossible possible. researchgate.net Cited by 104
[17] R Bal, IS Gill - 2020 - papers.ssrn.com. Policy approaches to artificial intelligence based technologies in China,
European Union and the United States. [HTML] Cited by 12
[18] K Foss-Solbrekk - Journal of Intellectual Property Law & Practice, 2021 - academic.oup.com. Three routes to
protecting AI systems and their algorithms under IP law: The good, the bad and the ugly. oup.com Cited by 26
[19] I Chatterjee - International Journal of Modern Research, 2021 - ijmore.co.in. Patenting machine-learning: review and
discussions. ijmore.co.in Cited by 7
[20] E Fok - Santa Clara J. Int'l L., 2021 - HeinOnline. Challenging the International Trend: The Case for Artificial
Intelligence Inventorship in the United States. scu.edu Cited by 11
[21] JC Trappey, CV Trappey, JL Wu… - Advanced Engineering …, 2020 - Elsevier. Intelligent compilation of patent
summaries using machine learning and natural language processing techniques. [HTML] Cited by 68
[22] A Trappey, CV Trappey, A Hsieh - Technological Forecasting and Social …, 2021 - Elsevier. An intelligent patent
recommender adopting machine learning approach for natural language processing: A case study for smart machinery
technology mining. [HTML] Cited by 36
[23] R Krestel, R Chikkamath, C Hewel, J Risch - World Patent Information, 2021 - Elsevier. A survey on deep learning
for patent analysis. uni-kiel.de Cited by 80
[24] J Yun, Y Geum - Computers & Industrial Engineering, 2020 - Elsevier. Automated classification of patents: A topic
modeling approach. [HTML] Cited by 55

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 70
Paper Publications
[25] S Baruffaldi, B van Beuzekom, H Dernis, D Harhoff… - 2020 - oecd-ilibrary.org. Identifying and measuring
developments in artificial intelligence: Making the impossible possible. researchgate.net Cited by 104
[26] S Sarica, J Luo, KL Wood - Expert Systems with Applications, 2020 - Elsevier. TechNet: Technology semantic
network based on patent data. [PDF] Cited by 149
[27] AV Giczy, NA Pairolero, AA Toole - The Journal of Technology Transfer, 2022 - Springer. Identifying artificial
intelligence (AI) invention: A novel AI patent dataset. [HTML] Cited by 51
[28] M Lee - World Patent Information, 2020 - Elsevier. An analysis of the effects of artificial intelligence on electric
vehicle technology innovation using patent data. [HTML] Cited by 45
[29] L Chen, S Xu, L Zhu, J Zhang, X Lei, G Yang - Scientometrics, 2020 - Springer. A deep learning based method for
extracting semantic information from patent documents. [HTML] Cited by 76
[30] J Wang, CC Hsu - Journal of Manufacturing Technology Management, 2021 - emerald.com. A topic-based patent
analytics approach for exploring technological trends in smart manufacturing. [HTML] Cited by 37
[31] Regenerate
[32] R Krestel, R Chikkamath, C Hewel, J Risch - World Patent Information, 2021 - Elsevier. A survey on deep learning
for patent analysis. uni-kiel.de Cited by 80
[33] DS Hain, R Jurowetzki, T Buchmann, P Wolf - … Forecasting and Social …, 2022 - Elsevier. A text-embedding-
based approach to measuring patent-to-patent technological similarity. academia.edu Cited by 37
[34] S Baruffaldi, B van Beuzekom, H Dernis, D Harhoff… - 2020 - oecd-ilibrary.org. Identifying and measuring
developments in artificial intelligence: Making the impossible possible. researchgate.net Cited by 104
[35] S Choi, H Lee, E Park, S Choi - Technological Forecasting and Social …, 2022 - Elsevier. Deep learning for patent
landscaping using transformer and graph embedding. [HTML] Cited by 25
[36] A Haghighian Roudsari, J Afshar, W Lee, S Lee - Scientometrics, 2022 - Springer. PatentNet: multi-label
classification of patent documents using deep learning based language understanding. springer.com Cited by 47
[37] AV Giczy, NA Pairolero, AA Toole - The Journal of Technology Transfer, 2022 - Springer. Identifying artificial
intelligence (AI) invention: A novel AI patent dataset. [HTML] Cited by 51
[38] L Liu, Y Li, Y Xiong, D Cavallucci - Computers in Industry, 2020 - Elsevier. A new function-based patent knowledge
retrieval tool for conceptual design of innovative products. academia.edu Cited by 78
[39] Y Luo, M Lin - International Journal of Intelligent Computing and …, 2021 - emerald.com. Flash translation layer: a
review and bibliometric analysis. [HTML] Cited by 34
[40] B Leten, S Kelchtermans, R Belderbos - Industry and Innovation, 2022 - Taylor & Francis. How does basic research
improve innovation performance in the world's major pharmaceutical firms?. maastrichtuniversity.nl Cited by 21
[41] J Dodge, M Sap, A Marasović, W Agnew… - arXiv preprint arXiv …, 2021 - arxiv.org. Documenting large webtext
corpora: A case study on the colossal clean crawled corpus. [PDF] Cited by 287
[42] A Bisoyi - Information Security Journal: A Global Perspective, 2022 - Taylor & Francis. Ownership, liability,
patentability, and creativity issues in artificial intelligence. [HTML]Cited by 10
[43] D Kim - GRUR International, 2020 - academic.oup.com. 'AI-Generated Inventions': Time to Get the Record
Straight?. oup.com Cited by 47
[44] T Tombekai - Available at SSRN 3772947, 2020 - papers.ssrn.com. The Ownership of Artificial Intelligence (AI)
Generated & Created Inventions. [HTML]
[45] F Mazzi - 2022 - qmro.qmul.ac.uk. Patentability of AI-generated inventions: a case study on pharma. qmul.ac.uk

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 71
Paper Publications
[46] L Mkhoyan - 2020 - dspace.aua.am. Authorship and ownership of intellectual property for objects created by artificial
intelligence.“can art and innovations created by artificial intelligence be protected …. aua.am
[47] SKA Dwivedi - RES MILITARIS, 2023 - resmilitaris.net. PATENT PROTECTION FOR INVENTIONS
RELATING TO ARTIFICIAL INTELLIGENCE (AI). resmilitaris.net
[48] S Chhavi, S Reeta - Legal Issues in the digital Age, 2020 - cyberleninka.ru. AI-Generated Inventions and IPR Policy
During the COVID-19 Pandemic. cyberleninka.ru Cited by 1
[49] MK Odeh - The Gravitas Review of Business & Property Law, 2020 - papers.ssrn.com. Patenting Inventions
Generated by Artificial Intelligence: The Way Forward. gravitasreview.com.ng Cited by 1
[50] R Matulionyte, JA Lee - SCRIPTed, 2022 - HeinOnline. Copyright in AI-generated works: Lessons from recent
developments in patent law. mq.edu.au Cited by 5
[51] EE Adaka, IA Olubiyi - J. Intell. Prop. & Info. Tech. L., 2022 - HeinOnline. Lessons for Nigeria: determining
authorship and inventorship of artificial intelligence generated works. strathmore.edu Cited by 1
[52] TW Dornis - European Intellectual Property Review (EIPR) …, 2021 - papers.ssrn.com. Of 'Authorless Works' and
'Inventions without Inventor'–The Muddy Waters of 'AI Autonomy'in Intellectual Property Doctrine. [HTML]
Cited by 9
[53] D Kim - GRUR International, 2020 - academic.oup.com. 'AI-Generated Inventions': Time to Get the Record
Straight?. oup.com Cited by 47
[54] AC Comer - NCJL & Tech., 2020 - HeinOnline. AI: Artificial Inventor or the Real Deal?. unc.edu Cited by 18
[55] Y Hao - J. Pat. & Trademark Off. Soc'y, 2024 - HeinOnline. The Rise of" Centaur" Inventors: How Patent Law Should
Adapt to the Challenge to Inventorship Doctrine by Human-AI Inventing Synergies. [HTML]
[56] E Bonadio, L McDonagh, P Dinev - Intellectual Property Quarterly, 2021 - papers.ssrn.com. Artificial intelligence as
inventor: exploring the consequences for patent law. gold.ac.uk Cited by 9
[57] KR Knutson - Cybaris Intell. Prop. L. Rev., 2020 - HeinOnline. Anything you can do, AI can't do better: An analysis
of conception as a requirement for patent inventorship and a rationale for excluding AI inventors. mitchellhamline.edu
Cited by 7
[58] DL Schwartz, M Rogers - Harv. JL & Tech., 2021 - HeinOnline. " Inventorless" Inventions? The Constitutional
Conundrum of AI-Produced Inventions. harvard.edu Cited by 14
[59] E Fok - Santa Clara J. Int'l L., 2021 - HeinOnline. Challenging the International Trend: The Case for Artificial
Intelligence Inventorship in the United States. scu.edu Cited by 11
[60] T Hauer - Humanities and Social Sciences Communications, 2022 - nature.com. Incompleteness of moral choice and
evolution towards fully autonomous AI. nature.com Cited by 8
[61] P Hutchinson - IEEE Transactions on Engineering …, 2020 - ieeexplore.ieee.org. Reinventing innovation
management: The impact of self-innovating artificial intelligence. [HTML] Cited by 69
[62] YK Dwivedi, L Hughes, E Ismagilova, G Aarts… - International Journal of …, 2021 - Elsevier. Artificial Intelligence
(AI): Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and
policy. openrepository.com Cited by 1812
[63] S Baruffaldi, B van Beuzekom, H Dernis, D Harhoff… - 2020 - oecd-ilibrary.org. Identifying and measuring
developments in artificial intelligence: Making the impossible possible. researchgate.net Cited by 104
[64] AV Giczy, NA Pairolero, AA Toole - The Journal of Technology Transfer, 2022 - Springer. Identifying artificial
intelligence (AI) invention: A novel AI patent dataset. [HTML] Cited by 51
[65] I Chatterjee - International Journal of Modern Research, 2021 - ijmore.co.in. Patenting machine-learning: review and
discussions. ijmore.co.in Cited by 7

ISSN 2349-7831
International Journal of Recent Research in Social Sciences and Humanities (IJRRSSH)
Vol. 11, Issue 2, pp: (63-72), Month: April - June 2024, Available at: www.paperpublications.org
Page | 72
Paper Publications
[66] HHN Abadi, M Pecht - IEEE Access, 2020 - ieeexplore.ieee.org. Artificial intelligence trends based on the patents
granted by the united states patent and trademark office. ieee.org Cited by 47
[67] R Righi, S Samoili, ML Cobo, MVP Baillet… - Telecommunications …, 2020 - Elsevier. The AI techno-economic
complex System: Worldwide landscape, thematic subdomains and technological collaborations. sciencedirect.com
Cited by 30
[68] I Kitsara - … and Artificial Intelligence: The Next Generation of …, 2022 - Springer. Artificial intelligence and the
digital divide: From an innovation perspective. [HTML] Cited by 26
[69] AJC Trappey, CV Trappey, JL Wu… - Advanced Engineering …, 2020 - Elsevier. Intelligent compilation of patent
summaries using machine learning and natural language processing techniques. [HTML] Cited by 68
[70] G Damioli, V Van Roy, D Vertesy - Eurasian Business Review, 2021 - Springer. The impact of artificial intelligence
on labor productivity. springer.com Cited by 140
[71] CH Yang - Research Policy, 2022 - Elsevier. How artificial intelligence technology affects productivity and
employment: firm-level evidence from Taiwan. [HTML] Cited by 118