- Biological Inventions
- Brand Valuation
- Competition Law
- Constitutional Law
- Consumer Law
- Copyright Infringement
- Copyright Litigation
- Corporate Law
- Digital Right Management
- Educational Conferences/ Seminar
- Fashion Law
- Hi Tech Patent Commercialisation
- Hi Tech Patent Litigation
- Intellectual Property
- Intellectual Property Protection
- IP Commercialization
- IP Licensing
- IP Litigation
- IP Practice in India
- IPAB Decisions
- Legal Issues
- Media & Entertainment Law
- News & Updates
- Patent Act
- Patent Commercialisation
- Patent Filing
- patent infringement
- Patent Licensing
- Patent Litigation
- Patent Marketing
- Patent Opposition
- Patent Rule Amendment
- Pharma- biotech- Patent Commercialisation
- Pharma/Biotech Patent Litigations
- Protection of SMEs
- Section 3(D)
- Social Media
- Sports Law
- Telecom Law
- Trademark Litigation
The introduction or coming into being of artificial intelligence and its development has formed challenges in the world of Intellectual Property law; especially copyright law. It is now similar to the situation when computers were introduced and at that time what was seen that whether the work that was made or generated by computer was protected under the wide amid of copyright law or not? The result which was logical that came out of this question or the answer to this very question was that computer might have produced the work but the work was created because of human touch or human influence by giving the computer commands to create the work. The computers were used merely as a tool to produce works where the creator saved himself time and labor to create such works. The computers cannot function on its own, it needs a human to operate.
Then came the artificial intelligence systems that brought about a storm in the world of intellectual property law. Since the year 2010 artificial intelligence has developed by tremendous speed. For example – Google’s Deepmind, what it does is that it is based on the concept of machine learning. Now, first of all, what do we mean by machine learning, well machine learning is a process where the machine on its own develops itself or improves itself by accessing data and use it to learn by themselves. This deep mind learns to play videogames similar to how human beings play. What it does is that it creates a neural network or as I would like to call it a mini-brain of itself and learn itself by accessing data from various sources. The work that artificial intelligence creates is no match to the work that a genius person creates but it may be better than what you and I can create that is it can be better than what an ordinary human being creates. The intelligence that is used in the works that are created by the artificial intelligence system is not that of a human being but of the artificial intelligence system itself. The works created by artificial intelligence is on par with that of human beings. Artificial intelligence generated work or created work satisfy the human needs as in what the humans demand therefore in my opinion they should be protected under the copyright law as they satisfy the interests of the public at large.
The laws that are prevalent in today’s world or as you can say the position of current copyright laws in many countries is that they protect only works generated by human beings, they do not protect works created or generated by machines. The works generated or created by artificial intelligence systems are not regarded as works therefore not subject to copyright. Now the big question that has arisen or raised is that should artificial intelligence generated works be protected or not? This question itself gives a view that the laws related to copyright law should be amended to include works created or generated by machines or Artificial Intelligence machines be protected under copyright law.
WHAT MAKES AI CREATIVE?
Artificial intelligent systems have grown at a tremendous speed from being a laboratory tech. To now being of some practical usage. In today’s time, artificial intelligence machines have come so far that now they are designed in such a way to produce useful work on its own. Some of its features are provided below:-1. Creativity
The artificial intelligence system not only copies stuff from sources that are accessible but also they create work that is original and new. They work as creative devices. This characteristic of artificial intelligence machines is very important in the world of intellectual property especially when copyrightable work is in question.2. Autonomous
If a device can make or do a high-level task on its own without any external help then it is said to be autonomous. They can work independently without or with minimum human influence. Artificial intelligence machines also make works autonomously and without any human intervention.3. Unpredictable
The artificial intelligence system can create new work without copying the original one as they are capable of creating unpredictable routes to optimal solutions. What it does is that it breaks up the data and then recomposes them to make a completely different work altogether.4. Learning Capability
The artificial intelligence system receives data and then it continues to process data because of the use of feedbacks it gets and keeps on improving the results with the help of those feedbacks.5. Efficiency
Artificial intelligent systems are accurate and can process large amounts of data accurately and efficiently than a human being is capable of. They are more efficient and accurate than humans themselves.6. Free Choice
Artificial intelligent systems have a free choice mechanism that means that if they have alternatives then they will choose the best alternative and arrive at a solution that is the best outcome.7. Goal-Oriented
The artificial intelligence systems are goal-oriented in the sense that they work according to the goals that they are set for. E.g.- story writing, drawings, etc.
CAN ARTIFICIAL INTELLIGENCE GENERATED WORKS BE PROTECTED UNDER CURRENT STATUTES?
Under the Copyright Law of Japan, copyrightable works are “production or works in which thoughts or sentiments are expressed creatively and which falls within the literary, scientific, artistic or musical domain.”
The term creativity here means that the personality of the author is expressed in any way this can be inferred by various court cases or legal precedent. Therefore, only works created by humans can be protected under copyright law. Therefore, the Copyright Law of Japan thus does not protect works created by artificially intelligent machines.
Under the Copyright Law of Germany, it says that “works in this law are only individuals’ intellectual creations” (section 2, paragraph 2)
This means that works created by only individuals are protectable and therefore the works created by artificial intelligent machines are not protectable under the Copyright Law of Germany.
Under the Copyright, Designs and Patents Act of UK it is provided under section 9 paragraph 3 of the act says about the author of the work: –
“In case of literary, dramatic, musical or artistic work which is computer-generated, the author shall be taken to be the person by whom the arrangements necessary for the creation of the work are undertaken”
So, under the Copyright, Designs and Patents Act of UK the works generated or created by the artificial intelligent machine are protected.
UNITED STATES OF AMERICA
United states constitution has the power to make congress enact copyright laws which say “the congress shall have the power to promote the progress of science and useful arts for limited times to author …. the exclusive right to their respective writing…” (Article 1, Section 8, Clause 8)
The United States supreme court said that in a case that “writings are founded in the creative powers of the mind”. There has been no case related to the artificial intelligent machine works and also the United States Copyright Office has expressly declared that the works created by artificial intelligent machines may not be protected. (The Compendium of US Copyright Office Practices)
THE ORETICAL JUSTIFICATIONS
The next big question is whether the theories of copyright allow for the protection of works generated or created by artificial intelligent machines? To answer this question there are some theories laid down below:-
The Labor Theory
This theory was given by John Locke in his work “Second Treatise of Government” as follows:
“Though the earth, and all inferior creatures, be common to all men, yet every man has a property in his person: this nobody has any right to but himself. The labor of his body, and the work of his hands, we may say, are properly his. Whatsoever then he removes out of the state that nature hath provided, and left it in, he hath mixed his labor with, and joined to it something that is his own, and thereby makes it his property.”
What this theory tries to explain is that when labor is done then the value is created and the person who has given birth to that thing should also have the right to enjoy it because of labor being put into by him. But this theory only talks about the labor of human beings and not of machines therefore it does not justify if the works created by artificial intelligent machines should be protected under copyright law or not.
The Personality Theory
This theory was given by Georg Wilhelm Friedrich Hegel in his work “philosophy of right” as follows: –
“A person has as his substantive end the right of putting his will into any and everything and thereby making it his because it has no such end in itself and derives its destiny and soul from his will. This is the absolute right of appropriation which man has over all things.”
What this theory tries to explain is that if a person has created a thing by putting his expression into that thing then he has a right over that property. Here the works created by the artificial intelligent machine are not derived from its own will. Therefore, the works created by the artificial intelligent machines cannot be considered as work protected under copyright law.
The Incentive Theory
The United States Constitution states that the purpose of copyright or aim of copyright is to promote the development of science and arts and not to protect the personality or labor of the creator or author. The United States Supreme Court laid down the incentive theory in a case Sony Corp. v. Universal City Studios, Inc., where it said that:-
“The monopoly privileges that Congress may authorize are neither unlimited nor primarily designed to provide a special private benefit. Rather, the limited grant is a means by which an important public purpose may be achieved. It is intended to motivate the creative activity of authors and inventors by the provision of a special reward, and to allow the public access to the products of their genius after the limited period of exclusive control has expired.”
This theory tries to say that copyright is used to protect the creation of the work for the public’s benefit and gives an incentive to authors, motivates them to create more and more creative works. So if work is created by artificial intelligence machine then it should be protected to promote creations of works like that. This theory justifies why works created by artificial intelligence machines should be protected under copyright.
PROBLEMS IN APPLYING THE TRADITIONAL APPROACH (AUTHORSHIP)
There comes a lot of obstacles in applying traditional copyright frameworks to works that are created by artificial intelligent machines.
Machine as Author
There is this possibility or proposal that we give authorship to the artificial intelligence machine or software. By giving allocation or authorship rights to the machine itself we will align with the court’s approach of awarding the person who is responsible for the creation of the copyrightable work. As the machine is responsible for the end product there comes a little difficulty that software and machines are not a legal entity. It is logical to grant authorship rights to the machine itself but the courts will not grant an entity the authorship rights which is not even a legal person or entity.
If we turn machines into authors then the only benefit that comes out of it is that it is based upon the very core idea of intellectual property rights that the creator is the author itself. This will also be beneficial because then the complex legal solutions to assign the authorship to the persons who have not contributed to the creative process will be solved.
It is also possible to assign authorship rights to the artificial intelligence system in the circumstances stated below: –
- That artificial intelligence machine creates works that are random or can’t be anticipated.
- The algorithm on which the artificial intelligence machine works should be devoid of any human intervention to ensure it acts independently.
- The artificial intelligence machine on its own to decide when to create the work.
Machines can be allocated authorship rights only when they act like humans and the characteristic of self-awareness develops into them. The machine should be capable of deciding when and how the work is to be produced only then they will act autonomously. The European Union commission is trying to amend laws to provide rights to machines or robots. Now the core idea behind intellectual property rights is to provide the incentive to the authors but in the case of machines being author, they do not need any incentive to create works. If we provide authorship to the machines then the other people who are involved in the creative process such as programmer and end-user will be deprived of their rights and they have also invested their time, money and labor in the creative process as well. In this case, a solution is there that we shall provide joint authorship to the programmer and the algorithm. But the problem here arises that both the author’s work is to be distinguishable but in a complex mechanism such as machine learning, it is next to impossible to determine what was the author’s contribution in the creation of the work. So this solution of providing joint authorship also is not applicable because it is very difficult to determine who contributed how much in a complex mechanism involved in the works created by the artificial intelligence machines.
Programmer as Author
The programmer is the one person who has poured out his heart and soul. They are the ones who understand the algorithm and are also very close to the creative process and are the biggest contributor in the creative process itself. They can also explain the algorithm better than anyone else because basically, it is their brainchild that is the algorithm. They have invested their labor and time into making such an algorithm that they should be rewarded for it by giving them the authorship rights to the works created by the artificial intelligence machine. Without a programmer into the equation and its creativity, the final product would have not been made possible at all.
By giving authorship rights to the programmer the core idea of the intellectual property rights gets highlighted which is that it will give incentives to them to further improve the artificial intelligence machine and also will increase the number of artificial intelligence works.
The proof of such argument can be seen in cases such as Nova Productions Ltd v Mazooma Games Ltd., in this case, the parties were electronic pool game creators. Here the court that is the UK Royal Court gave the authorship rights to the programmer as he was the creator of the infringed element of the video game. Another example of a similar case like that can be of the Atari Inc. v. North American Philips Consumer Elec. Corp., in this case, also the authorship was given to the programmer who was the creator of in-game images and display which were disputed.
If the copyright is provided or granted to the programmer of the algorithm then it would be an injustice to the programme or the machine itself as the machine on its own created the work and not the programmer. The machine acted autonomously and not the programmer directly contributed to the making of the work possible. Also, the programmers will be rewarded twice if we give authorship rights to the artificial intelligence created work i.e., one for the creation of artificial intelligence machines and other for the outputs that come out of the artificial intelligence machine.
User as Author
The user should be the one who should get the authorship rights because he is the one who uses the artificial intelligence machine as a tool to express his creative side. It is based upon the fact that the tool would have never existed if it wasn’t for the user in the first place. But this position would be difficult to anticipate when the user’s contribution is to a single touch of a button. If we give authorship rights to the user or the end-user of the programme then that will diminish any incentives for the programmers to create or improve the artificial intelligence machine. This position cannot be defended because it will be seen as free-riding on the efforts and labor of the programmer which is unfair in my opinion. It will have a negative impact on the programmers skill as they will not get any incentives to make new works and therefore innovation will get kicked off from the world of artificial development and the artificial intelligence will not grow far or will not develop as it should have been.
In my opinion, as a law student, it should be given in the public domain so that every person should be benefitted out of it. It will solve the problem of authorship because then the user of the program that is the person who has taken up the open license of the program will have to use his creativity and develop or make change to the open-sourced programme and this way the user and the programmer will be the same person and this will ultimately solve the problem of the authorship as well. Work generated by the artificial intelligence machine should be openly licensed.
Immediate Entry into the Public Domain
The most plausible or effective solution for all of this is that the work generated by the artificial intelligence machine should be provided to the public domain. This solution is practical and has very minor drawbacks or few drawbacks. This means that the author is the machine itself here. It is software that is not a legal entity and so no rights will be deprived off of it because there were no rights at all to begin with. Also, the programmer will get his code copyrighted and because of that it will receive adequate protection and will also be compensated for. Well, the end-user was deprived of copyrighting the work generated by the artificial intelligence machine still he can do modification to that and can create a work of his own and in that derivative work, he can get copyright in it. By providing the works generated by the artificial intelligence machine into the public domain the public will be benefitted because of the access to creative work without compromising or providing injustice to the content producers.
If this kind of framework is adopted then
the infringers or the alleged infringers have to prove with evidence that the
work was authored by the artificial intelligence machine and the protection of
copyright that has been provided to it is not valid at all. In that case, this
public domain framework will have a difficulty of enforcing false copyrights.
Author: Yashik Sahni, 5th year student of Rajiv Gandhi National University of Law, Patiala, intern at Khurana & Khurana, Advocates and IP Attorneys. In case of any queries please contact/write back to us at email@example.com.
The Copyright law of Japan, 1970, art. 2(1)(i)
Act on Copyright and Related Rights, 1965, s. 2(2) (Germany)
The Copyright, Designs, and Patents Act, 1988, s.9(3)
The Constitution of United States of America, 1787, art. I sec.8(8)
 trademark Cases 100 U.S. 82 (1879)
 John Locke, “The Second Treatise of Civil Government”, University of Colorado website, available at https://rintintin.colorado.edu/~vancecd/phil215/Locke.pdf (last accessed on 19th October 2019)
 John Carlin, Culture Ventures: Artistic Appropriation and Intellectual Property Law, Heinonline website, available at https://heinonline.org/HOL/LandingPage?handle=hein.journals/cjla13&div=10&id=&page=, (last accessed on 19th October, 2019)
 464 U.S. 417 (1984)
 Jeffrey Helewitz and Leah K. Edwards, Entertainment Law, Delmar Cengage Learning Publishers, United Kingdom, 2017
  EWCA Civ 219;  EMLR 427; 
 459 U.S. 880 (1982);