Introduction
AI or Artificial Intelligence has made quite a roar in the Internet Landscape around the globe. Each and everyone has been touched by its eminent presence in the form of Chatgpt, Siri, Alexa, Grammarly, Gemini-formerly Google Bard, etc. A noticeable outlook regarding the Personality, Authorship of AI was brought in with the infamous case Ankit Sahni Case, where for the first time, the Copyright Office of India recognized an artificial intelligent painting app “RAGHAV” as a co-author of an artistic work, however, the recognition was challenged and removed from its deemed title, leaving a stir in the world of technology.
In this blog, the author discusses the various facets of Artificial Intelligence and the discourse between granting personhood to AI.
AI and Legal Personhood
Before thinking about AI and its Personality, one should have a clear and concise knowledge of who the Indian Law recognizes as a Person. The concept of Personhood has been discussed in of the Indian Statutory Laws, Firstly, Section 2(26) of Bharitya Nyay Sanhita,2023 -formerly Section 11 of the Indian Penal Code, 1980, Secondly, Section 3(42) of the General Clauses Act,1897 both define a person as “Person includes any Company or Association or body of persons, whether incorporated or not”. Thirdly, Order 33, Rule 1 of the Civil Procedure Code, includes any company, association, or corporation as an individual, whether incorporated or not.
What is interesting about all these definitions is that all of them are inclusive definitions and not exhaustive, meaning that all these definitions include but are not limited to the keywords in the sections. The Same was clarified by the Supreme Court of India in SK Gupta v. KP Jain held that where in a definition clause, the word “include” is used, it is so done to enlarge the meaning of the words or phrases occurring in the body of the statute. The author also takes into account cases such as Sashikanth Vasudevbhai Patel v. State of Gujarat, and Karnataka Bank Ltd v. State of A.P. & Ors, which also states that the definition under these sections is inclusive in nature.
From these definitions, two major aspects can be extracted, firstly, that Artificial Intelligence can be a Juristic Person and secondly, that the concept of Personhood is dynamic i.e. it evolves with the change in society or according to the current societal norm. let’s discuss both these aspects one by one.
Whether AI is a Juristic Person
A juristic Person is not a natural person i.e. A juristic person is created, but rather, artificially created by a natural person, or a group of persons, Such Persons are recognized by Law in the form of Companies, Societies, etc., clearly proving that the Law has the power to create a juristic person, which brings the author to interpret that giving AI a Personality or granting AI personhood is, in fact, possible and probable in the near future.
One can approach personhood in one of two ways, one being breaking down AI to its basic structure and creating a nexus between how AI and a Human Brain are two different branches of a neural network that boils down to a similar pathway of conduction. And second, is the creation of a nexus between the concept of a company and AI and why the personhood of a company and AI are two roots of accountability and liability.
AI and Corporations
Firstly, let’s derive an analogy as to why corporations and companies were given the status of juristic persons, the first and foremost reason for the same was to hold the corporations liable, and accountable for their actions and decisions and to address the consequences of the company by the company and not on an individual’s shoulder. Another reason was to motivate people to engage in commercial activities with respect to the Industrial Revolution through the creation of corporate entities. In the same vein, the concept of legal personhood can be extended to Artificial Intelligence.
However, the Proportions of Liabilaity come bearing its facets of challenges and discourse. The major problem in extending AI personhood in the same vein as a corporate entity is that a corporation is made of Human beings through whom it operates, whereas an AI system is made by Humans, providing a clear contrast in the structure of both entities.
As an arguendo, let’s consider AI as a juristic person. Now major concern for the government would be to apportion the liability of AI. The following questions arise in such a state. Firstly. Who will take on the liability? There are three ways to answer such a question, one, is the responsibility of the creator of the AI, second, is the Owner of the AI, and third, is the person in command of the AI. The dilemma here is who does account for the liability, the creator who might have sold the AI. Or the owner who bought AI in good faith or the person who gave the prompt to AI to do an act that caused liability.
Consider the same with an example: A created an algorithm Self-Driving Car. B, a company bought out the AI for mass production. C, a person who bought the car, drives it. In case of an accident. C, says that he followed the rulebook to operate the car and did no such thing that could lead to an accident, and therefore the Car’s algorithm is faulty, and therefore B, the company is liable. B, says that I bought the algorithm on the assurance of A, who created an AI and gave a guarantee as to the proper functioning of the program, the same is also patented under his name. Therefore, extending liability to A. C, says that he created the basis of the driving system of the car, and B’s usage or the application of his software led to the accident. Creating a doom loop of liability without ever reaching a correct or one way of proportioning liability.
For a second let’s believe that the first question is tackled, even then the question of how liability can be traced Criminally and Civilly is a difficult issue. In a case, where AI accumulates wealth raises the question of how it might be taxed and to whom it ultimately belongs, and whether such property can be used to buy a property in the name of AI. Leading to whether AI can enter into contracts or sue or be sued. Criminally, how will AI be punished is a crater, considering the above example, the self-driving car due to a bug in its system, operates on its own and kills a person, how can one attribute liability in such a case? Before we answer the following questions the personality of AI is like a person balancing on a high held rope.
AI and Human brain
Artificial intelligence is based on Deep Learning. Deep learning is a technique for implementing Machine Learning. Deep Learning was inspired by the structure and function of the brain, specifically the interconnecting of many neurons. Artificial Neural Networks are algorithms that are based on the biological structure of the brain. In Artificial Neural Networks, there are ‘neurons’ which have discrete layers and connections to other “neurons”. Each layer picks out a specific feature to learn.
John Searle, an American Philosopher, calibrated the functioning of the Human Brain with that of digital computers in the view of philosophy, psychology, and Artificial Intelligence summated that a Strong AI is to Computer Hardware what the Mind is to the Human Brain.
However, a common objection to such an argument is that AI lacks Creativity, emotions, and Free Will as opposed to the Human Brain, the electrochemical pathways leading up to the actions and reactions of input might be similar in both the Brain and AI at a neural level but the same does not withhold the capacity of the human brain to feel emotions such as Happiness, Grief, Sadness, Joy, disgust, envy, empathy, etc. which isn’t the same for AI. AI cannot feel things, and cannot imagine the human brain, therefore, proving contrary to the concept of personhood.
A counter to this can be that AI and the Human Brain are associated with the ability to reciprocate and provide feedback in whatever form the two are comfortable. The AI can react or give feedback to the same situation in a logical, rather, rational, or objective approach by analyzing various options to a question where the human brain can be clouded with emotion. The inability of the AI to interact without emotions may not be a bigger challenge to overcome than we might think.
Personhood is a Dynamic Concept
Traditionally, personhood as a concept was only attributed to Natural Persons i.e. us, Human Beings. For centuries in time, Personhood was not only denied to millions of women, but it wasn’t even questioned as to why they were not attributed the same personhood as a man. But with the onset of feminist literature, women were granted personhood, making them their own person with the capability to vote, open a bank account in their names, buy a property in their names, and what today we see as a norm, was considered rudimentary in centuries past us.
A significant development in the concept of Personhood came with Christopher Stone’s rousing paper where he questioned the very essence of personhood, breaking down the concept into Three requisites, first, that it can institute legal actions at its behest; second, that in determining the granting of legal relief, the court must take an injury to it into account; and, third, that relief must run to the benefit of it. Stone’s paper questioning What if Nature and forests, could be subjected to legal Rights and Legal Personhood opened an entirely new array of ideas that could be incorporated as a person.
The earliest example of this shift comes from Aotearoa New Zealand, which granted legal personhood to the Te Urewera National Park, as part of a treaty dispute settlement with Māori. In 2017, Te Urewera was joined by Te Awa Tupua (the Whanganui River), and then by the Ganga and Yamuna Rivers in Uttarakhand, India, and the Río Atrato (Atrato River) in Colombia by 2020, Lake Erie in the USA had gained and lost legal rights, and all rivers in Bangladesh were recognized as legal persons and living entities.
In India, along with recognition of the rivers, in the Dakor Temple Case (1887) the court held that: “Hindu idol is a juridical subject and the pious idea that it embodies is given the status of a legal person. In another case, the court held that it is not all idols that will qualify for being a juristic person but only when it is consecrated and installed in a public place for the public at large. Given that AI is in the hands of the Public, available via a public forum, the author believes that there is a possibility to give AI personhood, which it so parallelly is tending to.
AI and Advancement of Technology
This author believes that having a conversation about AI and its personhood gives us enough time to ensure a gradual shift to our laws and current legal systems to partake in a smooth and harmonious technological transition in the law.
Granting AI personhood would also provide a strong influx of research and development in the field of Artificial intelligence. It would encourage innovation and draw investments from around the globe to create cutting-edge technology, resulting in benefits for human beings, and also would serve as a boon to our economy.
The Ministry of Electronics and Information Technology in its report of Committee-D Cyber Security, Safety, Legal and Ethical Issues discusses in great detail what possible challenges we might have to overcome in the coming age of Artificial Intelligence in a wide array of topics, including Privacy, Weaponization, warfare among others. Whilst the report doesn’t discuss the personhood of AI it deep dives into the complexity and intricacy of AI giving us insight into plausible challenges while awarding AI personhood.
Conclusion
The author surmises that it is indeed possible to grant AI personhood in the coming future, the strong armed AI is a very niche candidate to attribute personality, the concept of corporate shield, and piercing through the shield when proportioning liability makes it easier to consider AI as a juristic person, being able to sue and be sued. AI certainly does qualify Christopher Stone’s three requisites to legal personhood and it is time for us to venture into a new arena to tackle it before it becomes an imminent burden.
Authored By – Khushi (Guru Nanak Dev University, Amritsar)
