This article has been written by Safitha Rahim pursuing a Diploma in US Intellectual Property Law and Paralegal Studies at LawSikho and has been edited by Shashwat Kaushik.
It has been published by Rachit Garg.
Table of Contents
Introduction
Intellectual Property Rights (IPRs) are those intangible assets that are granted to the creator or inventor of a valuable creation of mind or invention that can be monetized and gain a certain reputation. Intellectual Property Rights include copyright, industrial properties like trademarks, patents, trade secrets, designs, and traditional rights. Intellectual property is an evolving yet very important aspect of the global money market.
Artificial Intelligence (AI) is a type of machine with human-like intelligence built from complex algorithms and mathematical functions. A machine will only become artificially intelligent if it has generalised learning skills, reasoning ability, and problem solving capacity. In the general sense, it is a programmed machine mimicking human behaviour. There are mainly three different categories of artificial intelligence. They are:
- Artificial Narrow Intelligence
This type of AI is considered weak. They are programmed to perform only specific Tasks. The best examples of this weak AI are Alexa, social humanoid Sophia, etc.
- Artificial General Intelligence
This type of AI is considered strong AI. They are programmed to perform any task that requires human intelligence.
- Artificial Superintelligence
This type of AI is hypothetical because it is not programmed in reality. It is calculated that this type of AI can surpass humans with its programmed intelligence.
Intellectual property and its interaction with artificial intelligence have created new works in the present time. Many controversies have been raised about it. The main dilemma is whether the machines are conquerors of human creativity or creators of new works.
Artificial Intelligence and copyright
Chapter 3, Section 13 of the Copyright Act of 1957 defines the work that will get copyright. Copyright law only recognises the creation of minds that are original; that is, they should have a human author.
Artificial Intelligence has been used in works of art since the 1970s. The Artificial Intelligence tools have been used and are currently being used at a more advanced level in writing literary works, journalism, music creation, drawing, and gaming. For example, Google owns an AI company named Google DeepMind. This company’s deep learning algorithm does not have any rules, but their motto is to learn. To be more specific with its creation, the computer software can generate music while listening to sound recordings.
India’s Copyright Act, 1957, expressly defines who is an author of a computer programme under Section 2(d)(6) and what is a computer programme under Section 2(ffc). The computer software is copyrighted under literary works, and this copyright can be protected for a period of 70 years after the death of the author. But whether the artistic work created by an AI is copyrighted under the respective law will be a topic of discussion. Recently, the US Copyright Office issued rules regarding the use of AI in newly created artistic works, stating that copyright will only be granted to the material part that a human created and not to AI.
In the landmark case of Infopaq International A/S vs. Danske Dagblades Forening (2009), the Court of Justice of the European Union held that a work generated from a computer programme can only be protected if the author’s creativity is included.
Even though the law does not expressly prohibit copyright for AI art works, the laws are not amenable to works lacking human creativity.
Who will be the owner
The core question regarding the new creative work generated by an artificially intelligent machine is that who is the owner of the work? Whether it is the programmer of the AI software, the person who is using the particular AI to create that work, or the AI itself. Generally, the author gets the copyright over the created work, but if the author is an AI software, then the generality cannot be followed. If the AI software is to be recognised as the author of the created work, then the AI machine should be granted the status and rights of a human.
At present, when AI generates creative work, usually that part will be excluded from the copyright, and only the human creativity in the work will get copyrighted. It is because AI is not actually producing new work; it is mimicking what it learns during its programming. Because the creative work that AI software makes will have a substantive similarity to what it learns, The introduction of ChatGPT into the writing world has compressed the complex writing process to a single click. Because AI software learns from millions of pieces of information and utilises the information when it creates new works. AI can function for a minimum of time and generate complex tasks that may be impossible for humans to do alone. The newly created masterpiece ‘The Next Rembrandt’ using cutting-edge information technology has astonished the world because AI has reclaimed the classic era of masterpieces.
Artificial Intelligence and trademarks
Section 2 (1)(zb) of the Trade Marks Act of 1999 defines the term trademark.
The trademarks, logos, and brands are the source of identity for the products or services offered by the particular company, trade, or commerce. Trademarks are an important aspect of the current economy because new businesses and start-ups are emerging each and every day. There are thousands of brands and thousands of creatives to support those brands. Thereby resulting in continuous growth in the number of registered trademarks and the number of applications for trademark registration.
Artificial Intelligence has a significant effect on the creation of a trade mark or logo. The trade mark can be well groomed, from its font style to the impression that can be left on a customer, with the help of AI software. But unlike copyright, here the ultimate idea of the respective mark will be the creation of a human mind, and the AI can make it more appealing to the eyes of the customers.
The important aspect of the AI software on trade marks is its efficiency to collect a broad level of data and analyse it within seconds, thereby saving human labour, money, and time. The AI tools can make better use of databases and search engines. Due to the increase in registered trademarks and trademark applications for registration, it will be hard for a human to scrutinise every single application and identify deceptively similar marks, identical goods, and similar services from them. Whereas an AI software can exploit the search engines to the fullest and thereby find deceptively similar and confusing products and services by comparing one mark with another, and even the AI software can assess to some extent whether the trademark will get approved by the registry for registration.
One of the landmark cases related to trademark infringement is Lush vs. Amazon (2014). In this case, it was held that Amazon used the Lush keyword on Google to achieve wrongful gain by utilising the reputation of the Lush cosmetics company, where Lush was not interested in selling their products on the Amazon e-commerce platform. Thereby, Amazon has infringed on the Lush trademark.
As far as trademarks are concerned, the success rate of the products and services in the market will mostly depend on the perception of the customers. The AI software can help with customer searches, recommendations, and suggestions. The AI can make the product recommendation by highlighting the product’s use, quality, and efficiency and thereby boost the interaction between the products and customers, but the ultimate discretion will be on the customers to decide whether to buy the product or not.
What if AI does not give desired result
![](https://blog.ipleaders.in/wp-content/uploads/2021/10/cyber-law.png)
The artificially intelligent machine can help with trademark searches and customer interaction to a great extent. But if the AI software does not give the exact search results, has a default in their analytical review, or does not notify similar marks, then it will be a problem for the person who is using it. Even though the AI software offers a global range of monitoring solutions for trademarks, how much can we trust it?
Artifical Intelligence and patent
A patent is a type of intellectual property that prevents a third-party from using, selling,or making the invention of the patent holder. The patent is for a period of 20 years, which can be renewed by payment of a prescribed fee. The invention should be subjected to the inventive steps under Section 2(1)(ja) of the Patents Act of 1970 and need to have some essential features if it wants to get patented. They are:
- Novelty
- Non-obviousness
- Industrial application
Section 3(k) of the Indian Patent Act of 1970 expressly prohibits granting a patent to an invention by an Artificial Intelligence. Even though there are software patents or business method patents related to technology, when it comes to the inventions made by Artificial Intelligence, it is an entirely different scenario.
Machine learning, deep learning, and artificial Intelligence have taken innovations to the next level. But what happens if the inventor is itself a machine? The patent has granted a monopoly to its inventor, so the inventor can commercially exploit the invention. Because the intellectual property laws provide recognition for the invention as a reward for the innovation of the inventor. Currently, well developed countries use AI software to stimulate their innovative abilities and use the AI to create inventions too. If these inventions are not patented due to the participation of AI in them, then what will happen to the money and time they invested in their creation? The reality is that AI is used to assist with the potential database and simulation for the invention. Human creativity as well as machine learning skills are combined and used to create innovation.
But there is an actual instance where the inventor of an innovation is an AI machine.
In the landmark case of DABUS (Device for the Autonomous Bootstrapping of Unified Sentience), created by Dr. Stephen Thaler, the applicant was the AI machine itself. The patent offices of the USA, England, and Australia have rejected the application on the basis that the inventor should be a natural person and not a machine. At last, for the first time in the history of intellectual property rights, an AI machine has been granted a patent from South Africa on July 28, 2021.
Artificial Intelligence and design
Section 2(d) defines the term design under The Designs Act of 2000. Designs are the functional aspects of a product that can make a relevant impact on the minds of customers. The designs can portray the aesthetic values of the products, thus forming relevant parts of them.
Artificial Intelligence has the ability to design not only logos but also web pages and apps. Artificial Intelligence’s introduction to the world of design has opened a new age of mass communication and the utilisation of diverse art styles and software, different from the decades of traditional ways of designing. With the help of AI, any person can create a design that depicts exactly what the designer intends to say through the design work. There are AI image generating creative platforms like DALL E 2 and MIDJOURNEY. These types of AI can generate realistic images and designs from simple descriptions in the natural language, that is, prompts. The prompt should not be complicated for the AI to perceive, and the description given can be in different styles. But the most insane feature of this AI software, like DALL E 2 is that it is free.
The most astonishing AI machine that looks like an emerging threat to designers is Uizard and their newly discovered creation, Uizard Autodesigner. They are also creative platforms that generate unique image prompts. But Uizard can even generate designs from a sketch or doodle drawn on a piece of paper and screenshots of an app. The Uizard Autodesigner is the next frontier of AI design and can generate original multi screen designs with a single text prompt. The possible question that will arise in the minds of readers will be, Whether AI will replace Designers? The answer is No, because human creativity and AI design are co-existing in nature. Actually, AI design acts as a catalyst for human creativity in design.
IP laws and AI in long run
Intellectual Property Rights should be subjected to flexible changes to accommodate the creation of Artificial Intelligence. In a decade, Artificial Intelligence will be present in every aspect of intellectual property. So the recognition of the work made by AI, protection of the work created, and liabilities to the AI if they infringe any right of another person and if they make any work that is against human morals and the law and order of the country are needed to be subjected to discussion. All of these will become reality if there is a human entity to represent the AI as a legal entity. A person who can be sued and sue on its behalf.
If AI creates more work and human creativity shrinks to the point where they have to just click a button to create an artistic work or an invention, that will be a serious problem. Human creativity will die out, and machines will take over. In the World Economic Forum’s The Future Of Jobs Report, 2020 has reported that, by 2025, AI will replace 85 million jobs worldwide and, at the same time, create 97 million jobs.
If AI software dominates, then human to human interaction will be seriously affected. Still, a controlled use of AI in invention and creation co-existing with humans can do wonders. They can upgrade our standard of living, save time and money, increase the efficiency of invention, and give new horizons to the creation of minds and other intellectual properties.
Conclusion
Artificial Intelligence is both terrifying and fascinating. AI has vast and deep influence in the area of intellectual property. The creation of mind and inventions made by artificial Intelligence are not protected under IP laws in the normal sense, but they can be if the overall AI machine is disassembled and examined. That the AI machine should be differentiated into respective parts like core, technology, and technical aspects. The core of AI, which is usually a complex algorithm or mathematical function, should be kept hidden as a trade secret, whereas technology can be exposed to the public, thereby allowing new innovations to be made. The technical use of Artificial Intelligence has been recognised and patented in countries like the United States, China, and Europe.
In this continuously developing world, AI inventions need to be protected on a broader level with suitable and strong laws across all countries because of their influence on large scale businesses, startups, and even individual innovations. For example, let’s look at a scenario where a gaming company invests millions for the music creation of the game with AI, but it cannot be protected under Copyright law, and anyone can use it without payment. It will be an injustice to that creation. The recognition of the rights of an AI machine for its creation and invention is debatable, but it is not impossible.
References
- The Intellectual Property Laws: 2023 Edition
- https://www.wipo.int/tech_trends/en/artificial_intelligence/story.html
- https://www.iiprd.com/computer-programme-infringement-of-copyright-act-1957/
- https://carnegiemnh.org/biomimicry-is-real-world-inspiration/
- https://spoor.com/dabus-the-rise-of-the-inventive-machines/#:~:text=On%2028%20July%202021%2C%20the,the%20patent%20by%20the%20CIPC.
- https://www.epo.org/news-events/in-focus/ict/artificial-intelligence.html
- https://www.toptal.com/designers/product-design/infographic-ai-in-design
- https://www.imf.org/en/Publications/fandd/issues/2020/12/WEF-future-of-jobs-report-2020-zahidi
- https://www.nature.com/articles/d41586-018-05267-x
- https://www.americanbar.org/groups/intellectual_property_law/publications/landslide/2018-19/january-february/pros-pitfalls-artificial-intelligence-ip-broader-legal-profession/
Students of Lawsikho courses regularly produce writing assignments and work on practical exercises as a part of their coursework and develop themselves in real-life practical skills.
LawSikho has created a telegram group for exchanging legal knowledge, referrals, and various opportunities. You can click on this link and join:
Follow us on Instagram and subscribe to our YouTube channel for more amazing legal content.
![](https://blog.ipleaders.in/wp-content/uploads/2021/01/Internship-Banner.jpg)