ABSTRACT
The ever-advancing technologies have led to the evolution of the concept of Artificial Intelligence. AI can be said to be human in the form of technology. It is the custodian of human skills and knowledge but without the moral or emotional side of the humans. AI has come in vogue due to the very fact of the laziness of humans due to which they do not want to do their tasks on their own, and thus have developed the technology to take their shape and place. Various efforts have been made to give a definitive meaning to the term, however, the concept being so dynamic, cannot be limited to fixed limitations of words and phrases. It asserts a different meaning with every use, in different aspects. Although it is the transfer of human skill and knowledge to technology, it cannot be given a place equal to that of a human, with the very reason that it lacks moral and emotional values and thus uncompetitive to make value judgments and informed decisions. However, ever since its evolution, there have been various developments in this field. It has been introduced in our day to day activities in various ways. The examples of the same are Siri, Alexa, etc, and our very first human-robot, Sophia. Keeping the advantages of AI on one side, various disadvantages cannot be isolated. Introducing AI in the field of law is a very challenging task and whether it will be a success or not is what this paper discusses. It also discusses whether an AI can replace lawyers and how can it overall benefit the legal industry.
I. INTRODUCTION
Artificial intelligence is the intelligence of humans incorporated in the technology to supplement or substitute the efforts of humans. Every other thing nowadays is instilled with artificial intelligence, doing such things, which at one point in time was not even possible by human beings. Intelligence is a system within humans by which one can have and develop creative ideas, which is now being artificially mechanized. The concept of AI may seem to be very complicated and impenetrable to those not conversant with technology and related aspect, but the concept is no longer alien to most of the sectors. The term, in common practice, seems to be synonyms with robots and automatons, but in reality, AI is much more than that. With the growth of AI in every sector, the legal industry is not much behind. The present Chief Justice of India, Justice S.A. Bobde has been yearning to incorporate AI in the working of courts in various ways. However, believing only upon the miracles done by AI, one cannot ignore the very downside of it.
II. ARTIFICIAL INTELLIGENCE: DEFINED
Artificial intelligence, in the simplest words, can be described as the transfer of human skill and knowledge to a machine to perform the day to day work of a human being to make life easier, but they lack an emotional capacity. There are, however, some accurate descriptions of what is artificial intelligence by various scholars.
According to Jake Frankfield “Man-made brainpower alludes to the re-enactment of human insight in machines that are modified to think like people and copy their activities. The term may likewise be applied to any machine that shows qualities related to a human personality, for example, learning and critical thinking.”[1]
The first definition of Artificial intelligence was coined by John McCarthy in 1956. It was the time when he had invited a group of researchers from various fields and backgrounds which also included language simulation, complexity theory, neuron nets and more to a summer workshop that was called the Dartmouth Summer Research Project on AI. It was done for discussion on what would finally become the field of AI.[2]
According to the modern dictionary, the definition of AI focuses on AI being a subdivision of computer science and how machines can imitate human intelligence the exact definition of AI “The theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.”[3]
Artificial intelligence definition is not something that is fixed and solid which can be used every time one mentions AI but it is very dynamic and keeps on changing and adjusting depending on its use. There have been companies that have adopted AI for various purposes like, Google that uses AI for machine learning and taking their photos on their smartphone. Other examples include Apple with Siri and Amazon who uses Alexa as their AI. The advantage of AI can be taken but it is the people who need to know how to take proper advantage and put AI for good use. There has been tendency of humans to misuse anything and everything. One such example where AI has been misused is with Facebook using personal information and infringing on their privacy.
III. BEYOND THE VERNACULAR: MEANING OF AI FOR THE LEGAL INDUSTRY
As Artificial intelligence is no longer a science fiction for the legal industry, various law firms have adopted AI in their daily task and have also defined what AI means in the field of law.
According to Westlaw and their developers, AI means to make the computers learn human behavior or one can say to mimic their behavior to find the most suitable and accurate results and solution for the problems or question that one needs an answer to save the time of the lawyers (legal professionals) without sacrificing quality. One of the better examples is the use of an AI search engine by the Westlaw developers that is the Natural language Search in Westlaw. In the early 1990s, it was very important to use Boolean Terms and Connectors to get a meaningful search result. But with the use of Natural Language search, it allowed to machine to understand the way humans thought. Thus, the introduction of Natural Language in Westlaw means that instead of the users to think like a machine it was the machine that started thinking like the users. This finally resulted in easier and faster search results.[4]
The biggest difficulty in producing a legal definition for AI is that the AI experts also do not have a fixed definition for it.
Five elements are integral to the current AI– Natural language processing, artificial neural networks, deep learning, search algorithms, and machine learning.[5]
The question arises that why is it important to define artificial intelligence in terms of the law. The answer lies in the fact that one needs to regulate AI. For instance, if a car driven by AI causes an accident then the person affected by it needs to know whom to sue for the damages caused. One of the most important things, to give a legal definition, is to make sure that the definition is future-proofed to accommodate any improvement or changes that may be brought to AI. There have been various attempts to give AI a legal definition. In the UK, a report released by the House of Lords committee provided:
‘Technologies with the ability to perform tasks that would otherwise require human intelligence, such as visual perception, speech recognition, and language translation.’
This definition was considered to be not so accurate because it talked about human intelligence which in itself is very difficult to define. Another definition was given by EU commission:
‘Systems that display intelligent behavior by analysing their environment and taking actions – with some degree of autonomy – to achieve specific goals.’
- LEGAL STATUS OF ARTIFICIAL INTELLIGENCE
There are new markets that are formed arising out of the technological paradigm of the digital economy. This, in turn, is giving rise to a certain subject of control which also includes Artificial Intelligence. The problem that arises out of the above trend is that the technology that arises out of it may force to displace the market economy forcing professionals out of their jobs thus causing unemployment. The situation for lawyers offers very little hope as there are already technologies that displace lawyers from the market. One such instance of technology substituting the tasks of lawyers is the DoNotPay chat in the United Kingdom and it has covered around one thousand fields of law. This technology came into the limelight because it had challenged around one lakh sixty thousand illegal parking tickets that were issued to the owners of the car and this challenge was successful.[6] Sberbank, in the country Russia had launched a robot lawyer and the purpose of this robot lawyer was to file the claims for the individuals. Similarly, a company had launched a bot and the purpose of this bot was to settle a dispute arising out of insurance. Thus, it was believed that to curb the advancing technological trend one needs legislation to regulate them. This issue is being looked at closely by both the researchers and other professionals in the legal field.
After looking to all the issues and the advances in the technological front in the field of law it was held and believed by professionals that a significant amount of pressure must be made to the issue of dealing with artificial intelligence as a different subject in control with administrative and civil and even criminal proceeding within that context that AI will be competing for the human beings.
There are generally two methods to implement the AI; first, it is either implemented in the form software, for example, Siri, Alexa, chatbots or virtual reality and the form for the implementation of AI is by program, for example, one can say drones and robots. The difference between a software AI and a programmed AI is that the software AI does not have a material shell. Some cases represent that the AI status is in contradiction with legal national norms:
Saudi Arabia case:
It was announced in the year 2017 by Riyadh the there is a robot named Sofia that considers itself as a robot that has been granted within the citizenship of Saudi Arabia. This, in turn, had contradicted various legal norms and conditions of Saudi Arabia. The first norm that it contradicts in the norm of Saudi Arabia citizenship which is given or obtained in the following ways: Firstly it can be obtained by birth in which it is stated that the birth must be in a traditional family where the mother and the father must be of Saudi Arabia citizens secondly there must be a birth of one of the legal entity in Saudi Arabia which means a family where the father was born in Saudi Arabia and is a citizen but the mother is not a citizen. Thirdly where the birth of the subject where the mother is the national and father is not provided that the subject must have a permanent address and must be fluid with Arabic.
The second way to obtain a citizenship in Saudi Arabia is by marriage and there are many other ways which include one must reach a certain legal age another is fluency in Arabic and settlement over 10 years one to have citizenship must have no criminal record, source of earning as well which must be continuous and the person must comply by the norms of the country. Furthermore, the case of Sofia the AI robot contradicts the accepted norms in which a female of Saudi Arabia must comply with. The norms that a women in Saudi Arabia need to follow include that they must move around accompanied by a man known as the male mahram, one needs to also wear a hijab there are further limitation regarding taking job for a female as well as their limits set which include limits in traveling abroad, limits in family including inheritance.
After looking at all the limitations one can say that the robot violates all the limits set for a female in Saudi Arabia.
Japanese case:
In the country of Japan in the year 2017 japan provided with a residence permit to a chatbot which was name Shibuya Mirai under a special regulation.
However, this causes and contradicts with the permit procedure that has been set in Japan. There are a set of people who get the permit for a residence in japan which include participants in the family reunification program, businessman and entrepreneurs doing their business in Japan, foreign specialist who have been employed in Japanese companies, athletes of world levels, artists of world level and recognition, people who have married a Japanese citizen, a foreign student who studies in a Japanese college for a higher education.
There are only two options to get a permit for residence either the person must be born in japan that is by birth or naturalization by naturalization means the person must have lived in Japan for more than five years and can reach the age of twenty and the person must have no connection groups and activities that have an objective against Japan.
After seeing all the above points that one needs to fulfil to get a permit for residence none of those points were achieved by the AI Shibuya Mirai as neither the chatbot was born in the nation of Japan neither it had met the criteria of naturalization. This is one of the reasons and the main one why AI cannot replace a human being as it does not have rights and duties as a human being.
- ARTIFICIAL INTELLIGENCE: EQUAL TO A PERSON?
After focusing on the above-mentioned cases of the AI status are, they equal to human beings in Saudi and Japan one must not shift their focus on the procedural aspect of the AI that means the public legal relations. One must question what does this public legal relation means.
Neither the chatbot or the robot Sofia had applied for the permit of residence neither the chatbot or Sofia meets the qualification for example age qualifications necessary. These were just a few things some major thing the chatbot and the robot had not met was that neither of them spoke the national language to the extent the law wanted its citizens to speak.
Due to the criteria not met by the chatbot or the robot both of them should have met with a refusal while they might have submitted the documents for citizenship or the permit of residence. Sophia the robot must face serious problems arising due to contradiction from the legislation of the host country. Where one female must have a male escort or have a limit in job, Sophia the robot did not comply with any of those criteria and thus must be brought to a criminal and administrative under the legislation that is currently in force in Saudi Arabia. The chatbot, on the other hand, has lesser responsibilities for contradicting the rules and regulation of the nation and the mere reason for this is that the chatbot had no material shell thus the complications were less. One can say that the chat bought being a less developed AI had fewer complications and problems compared to the robot Sofia. Another reason which causes less complication for chatbot was that it further tied to the location of the server.
One must in the legal field of studies keep in mind that along with the duties AI will receive rights as any living human being who is a citizen of a nation. In the case of robot Sofia, one must consider that if one compares Sofia with a human being then one must also know that females in Saudi Arabia now can participate in executive branches, marry and also take part in labour relations. Thus, one must see if the robot is compared to a human being there will be problems and questions arising in the court of Sharia and also in general jurisdiction, as there is no model of conduct that has been informed to the legal professionals.
The chatbot which is an AI performs legal function as well as acts as an operational intermediary. It acts as an intermediary between the government and the Shibuya community. If one says that the AI has got a permit of residence the question arises about the format of employment contract because that is the only way to get a residence permit.
If one does not give due care and attention this would lead to a legal consequence because the one would not know who hold responsible for any breach of law as the chatbot has no material shell and Microsoft the developer of the AI has itself insured by the fact that the AI is a person thus Microsoft face no consequences. One must also see the difference between a human being and AI as if one compares the moral or ethical aspect the AI, I have none compared to a human being
- AI AS A SUBJECT OF LAW IN DIFFERENT COUNTRIES:
As of keeping in mind the contradictions that may be caused by the AI many countries are very quick to create the necessary framework one might require to resolve the issue of regulating or keeping in check the AI as a newly formed subject of law.
The examples of the countries doing the same are as follows: United Kingdom formed a committee for AI in the House of Lords. The united states of America believe that the government does not think AI has a separate legal status as an individual but rather focuses on the definition of AI. The AI definition is providing in the bill on AI under section 3. As per the European countries there take on AI is that they pay huge attention to the legal regulation of the unarmed vehicles. One of the examples of this is that the German traffic act imposes any responsibility of an automated or semi-automated vehicle on the owner of the vehicle.
The biggest issue with holding AI responsible like one would hold a human being responsible today in legislation is the control of the over the third person if for example a breach of law is committed by the AI one has conflict regarding who to hold responsible either the one who owns the AI or the software developer that created the AI or the AI itself. EU had raised one of the better questions if the damage is caused by the AI who does one receive compensation from or is it the sole mistake of the AI or the fault is of the third person who might have programmed the algorithm of the AI. One special reservation is made about the AI that as they have a self-learning procedure one can decide in an event to whom to hold responsible for the breach. Russia has the same situation as the EU bill called the Grishin Law. The Grishin law imposes all the responsibility of the AI robot on the robot’s operator, developer, etc.
After reading the ways the countries have adopted to deal with the legal consequences on the AI it is common in Russia and EU that the AI is not held responsible but all the burden of the breach of law by Ai is on the owner of the AI. There are various challenging factors as well for example the EU does not touch the possible scope of the robot’s application in society.
VII. CAN AI REPLACE LAWYERS?
The biggest question that arises after discussing the legal status of artificial intelligence is whether AI can replace lawyers or put them out of business. The question that is AI being used in the legal platform then the answer is yes, it is being used widely by the legal professionals very diligently and effectively in day to day research works, etc. However, there are many experts as well as lawyers who believe that not only will AI have a huge impact in the legal field but they will also eliminate many legal professionals and take their place.
According to one of the civil litigators Tom Girardi, one day not using AI would be considered legal malpractice. According to him, it would be considered a shame for the lawyer to do everything on their own in the twentieth century when one can do everything on a computer.[7]
There are various reasons in the legal industry to believe that AI would boom in that field some of the examples for the same are as follows; right now, in the present time, the legal system relies on paralegal armies and various researchers to find out whatever information one needs to find out. One of the disadvantages of armies of paralegal and researchers is that they increase the expense of the firm as both the paralegals and the researchers cost a lot of money this is one of the areas where if the firm used AI, they would save cost significantly. Another area where the AI will be more useful is in place of the investigators in the office of public defender where an investigator can only interview each client for very small time that means this service is not only expensive but also inefficient. If one uses AI it would not only save time in researching but also speed up the judicial procedure, it would also reduce the burden on the court and the lawyers and firms as well. One research has shown that people are more comfortable disclose information with a machine rather than a human because people have an opinion that machines won’t judge them as the humans do thus this is another reason and field where AI would succeed that is where one needs to take interviews from the client.
There are many instances in the legal field where AI cannot perform functions but then there are situations where they would perform better than any lawyer for example when one needs to communicate with expert witness the AI would have all the information calculated beforehand like what was the earlier statement of the witness is there any variation or what was the comment of the judge. Now the question which many would ask is whether the lawyers would be afraid to lose their jobs if AI is employed. The answer is no because the job of the lawyer is to find a solution and do research and AI would help them perform these functions in a better, faster and efficient manner this would help them thrive in a hectic firm environment. According to Girardi even though the lawyers who take the help of AI would have fewer hours of work but that doesn’t mean they will start losing out work. AI would only enhance the effectiveness and speed of the lawyer and this in return means increased business and repeated clients will be there. If a lawyer who can do the job in less time using AI compared to a lawyer not using AI It would benefit him by speeding his process getting even more clients. This indicates for the law firms that they might get more revenue and increase in their client roster. Well, the firms which won’t adopt AI will have a competitive disadvantage. Now the question shifts to whether AI can replace judges or the court the answer I not completely but yes, they can for certain practices, for example, AI can be used for the backlogged cases and here the AI can help in providing a faster way of dealing with cases thus removing backlogged cases. Whether it is India or any other place all the courts suffer from backlogged cases and the solution for it is AI. Some courts are of the thought that AI should be used to determine whether a person is fit for bail or not by analysing his behavior which was earlier done by judges, this would only speed up the process and reduce some burden from the judges. Girardi is also of the belief that the role of AI won’t be limited only to the above-mentioned task of the belief that AI one day might be used in the court to decide certain judgments made by the clients like whether a doctor operated late? Whether the operation was done negligently? These questions are expected to be answered by the AI which would further decide whether such matters are worthy enough to go to trial or not ultimately saving time and being efficient for the courts and helping in focusing on major cases that require attention.
The further belief of Girardi is that AI should be used to judge a juror whether he is eligible or not, whether he was earlier a juror in any other case, if yes what was the judgment in that case, whether he was involved in any other accident or not? further AI could also be used to identify the what does the juror think upon an issue raised in the court AI can do this by seeing is there is change in temperature of the body or the body movement or the change in skin color this would help one know what does the jury think even before he expresses it. In conclusion, this data would only help in finding the right jury which would in turn help in the fair result.
VII. AI NOT A THREAT BUT A HELP FOR LAWYERS
Artificial intelligence, machines learning are some of the tools that have already arrived in big and established law firms. There have been significant advances that have been made in the field of AI and the lawyers are of the fear that their job might be in danger. Somewhere this fear is very natural because if one imagines AI would be able to accomplish the same job of the lawyer within less time and involving less cost. It is clear when a client sees that his case will cost less and completed faster they would choose AI over a lawyer but this theory is not as simple as it looks what according to me is the future is that the AI and lawyers will have to work hand in hand to achieve efficiency one desires.
One is of the wrong opinion that AI would replace lawyers whereas the truth is that AI will facilitate growth and also increase the productivity of the legal firms this all will be achieved by accuracy and efficiency. AI and its algorithms have already been used by the corporate industry and have started to bring change by due diligence, e-discovery, and management of the contract.[8] Activities that are considered “less routine” are also becoming easier to perform for example mining of the documents, legal research and predicting the case outcomes have become faster, easier, and cheaper with the use of AI software in the legal industry. It is believed that law firms and lawyers who will adopt AI will be considered to more sophisticated as they will achieve their goals and objectives more efficiently and effectively. Therefore, not only they will have an edge over lawyers who don’t know how to use AI but also will be preferred by the customers. Further law firms would also be forced to and give preference to those lawyers who know how to use AI algorithms to help the firm in benefiting.
The opinion of the legal industry is that AI won’t replace lawyers but lawyers who are using will replace the other lawyers who are not using AI.
According to Altman Wells survey, it depicts that the legal market is facing price competition, delay in providing service to its client, increase in new small independent firms thus automatically causing increase in the competition thus one has to believe the tradition law firms must understand the benefits of AI because if they don’t they will start falling behind the competition. It is not the sole responsibility of the law firm but also for the lawyers to embrace the need for AI because it will only help their performance. McKinsey Global Institute found that around fifty percent of the whole task can be automated whereas only five percent of jobs can be automated and it also found out through its survey that twenty-three percent of the lawyer’s job can be automated. It believes by the people mostly by the people working in legal industry that the advancing technology is going take place in the legal field but it is also believed that the top lawyers would all be performing the main work, with main work one means the top lawyers will be performing tasks which require higher skills whereas technology is expected to perform lower task mostly routine work.
One of the biggest difference between an AI and a human or one can say human having advantage over AI is the fact that AI can predict things based on the algorithms but it can never even after development match the human judgment .it is very simple and clear as machine learning improves and changes and adapts itself the human prediction will go down as one knows the machine predictions will be far superior and fast due to advance algorithms but that does not mean lawyers will go out of business because when a person combines the AI predictions with human judgment then it will be most ideal situation for a lawyer. The law firms that adopt AI will be able to provide real-time insights along with improved decision making along with that the lawyers will keep on working on things the technology cannot replace.
This makes us come back to our original definition of what AI was: Artificial intelligence is the transfer of human skill and knowledge to a machine to perform the day to day work of a human being to make life easier, but they lack an emotional capacity. This is a definition crafted by me but there are some accurate descriptions of what is artificial intelligence.
According to Jake Frankfield “Man-made brainpower alludes to the re-enactment of human insight in machines that are modified to think like people and copy their activities. The term may likewise be applied to any machine that shows qualities related to a human personality, for example, learning and critical thinking.”[9]
It is thus clear that AI legal status is not as equal to a human being and neither they are replacing lawyers not even in the distant future and the reason for the same is that AI cannot have human judgments or rights as a human as it does not complete all the requirements and thus AI and lawyers need to work together to achieve excellence.
In fact, in one of the recent statements of the CJI, S.A. Bobde, he has expressed his keenness on employing AI in judiciary that reads one million characters per second. The purpose is to help the judges get the gist of voluminous petitions quickly and increase the pace of case disposal, which is the need of the hour, seeing the pile of cases pending in our law courts.[10]
VIII. CONCLUSION AND SUGGESTIONS
Artificial Intelligence can be said to be the transfer of human skills and knowledge to a machine to perform the day to day work of a human being to make life easier, but they lack an emotional capacity. In the course of various discussions relating to equalizing of rights of both the AI and the human, it can be concluded it is very difficult and the reason behind it is the current infrastructure of the legislation and a lacking mechanism to control or equalize the right. Thus, if an incident takes place whether a criminal or any other incident it will be the duty of the executive bodies to decide the corpus delicti.
Furthermore, it can be concluded that the legislation believes that the public authority function can be implemented by AI. But here is a catch that AI might become a threat with self-learning but also an AI lacks the quality of value judgment as a human being and does not have emotions, moral or ethical values which is very necessary to be possessed by a civil servant or one dealing situations as faced by a lawyer.
Thus, the question as to whether AI can share the same rights as a human being can be answered in both affirmative as well as negative. A country must introduce an institution of self-regulation in the field of AI, which helps in developing standards along with codes that must be followed by the AI along with the owners of AI and to makes these codes mandatory to be followed. These institutions may be introduced as authorized bodies at a national level and not constrain it to a local level in the field of AI which would further help in deciding what activities are acceptable by AI and which are a threat to the society.
The consideration as to whether AI will replace lawyers, it can be concluded that it would not. However, the lawyers must adopt AI and not only lawyers but the legal firms as well because it will only give one-person edge over the other who has not adopted. It is very important for the owner and AI to function together to achieve the most out of anything. But looking into the present Indian scenario, where there is a manual recording of evidences etc by the law courts, adoption of AI in the everyday business of law, is still a long way to go and cannot be expected to be witnessed in a near future.
[1]Jake Frankenfield, Artificial Intelligence (AI), INVESTOPEDIA (Jun. 13, 2019, 7:32 PM), https://www.investopedia.com/terms/a/artificial-intelligence-ai.asp.
[2]Bornard Marr, The Key Definition of Artificial Intelligence That Explain Its Importance, FORBES, (Feb. 14, 2018,01:27 AM),https://www.forbes.com/sites/bernardmarr/2018/02/14/the-key-definitions-of-artificial-intelligence-ai-that-explain-its-importance/#3abb989e4f5d
[3] Id.
[4]Mary Schlaphoff, The meaning of artificial intelligence for legal researchers, THOMSON REUTERS, (Feb. 5, 2019, 3:08 PM), https://legal.thomsonreuters.com/en/insights/articles/meaning-of-ai-for-the-legal-industry.
[5]Sam Hancock, AI technology: A lawyer’s guide, LEXOLOGY, (Mar. 6, 2019, 5:00 PM), https://www.lexology.com/library/detail.aspx?g=db175813-0f67-4949-834e-cec0f8843605
[6]Atabekov, O. Yastrebov, Legal status of artificial intelligence across countries: Legislation on the Move, XX1 ERSJ. Pg. 773-782 (2018).
[7]Neil Sahota, Will AI put lawyers out of business, FORBES, (Feb. 9, 2019, 10:43 PM), https://www.forbes.com/sites/cognitiveworld/2019/02/09/will-a-i-put-lawyers-out-of-business/#295436731f00
[8]Todd, Artificial intelligence won’t replace lawyers- it will free them, LAW TECHNOLOGY, (Feb. 27, 2018, 7:00PM),https://www.lawtechnologytoday.org/2018/02/artificial-intelligence-wont-replace-lawyers-it-will-free-them/
[9]Jake Frankenfield, Artificial Intelligence (AI), INVESTOPEDIA (Jun. 13, 2019, 7:32 PM), https://www.investopedia.com/terms/a/artificial-intelligence-ai.asp.
[10]Excessive Tax Social Injustice by Govt: CJI, TOI, Jan 25, 2020, at pg. 10.
