Abstract
The technological advancements are changing so rapidly that the constant need to keep pace with the ongoing developments becomes the very need for survival. Since the creation of Artificial Intelligence, there has been seen a strong force to introduce this system in each and every field. This force has also reached the doors of Arbitration where the daunting question of AI replacing humans comes into the picture. This research paper aims to analyse the various opportunities and challenges caused due to AI in Arbitration and answer whether AI can really replace human arbitrators or not. The scholar has adopted the doctrinal method of study where it has analysed articles, journals, surveys, case studies and websites. The paper contains detailed information and explanation about AI and Arbitration, Advantages, Challenges and Suggestions on the topic. At the end of the paper, the reader will be able to understand both sides of the coin and will be able to make their own judgment regarding the replacement of humans with AI arbitrators.
Keywords
Artificial Intelligence, Arbitration, AI Arbitrator, ADR system, Unemployment
Introduction
The concept of Alternative Dispute Resolution (ADR) was introduced with the object to lessen the burden of the judiciary which is mainly flooded by piles of cases every day notwithstanding the unavailability of enough judges in the courts. Arbitration is one of the strongest pillars of the ADR system which strives to give quick justice outside the traditional litigation practice. The term arbitration is derived from the Latin word ‘arbitrari’ means “to judge”. In simple layman’s words, arbitration is conducted by a human arbitrator who acts like a civil judge and mainly gives a decision that is binding on the disputed parties. The Role of Artificial Intelligence (AI) in the Field of Arbitration has a significant spot to revolutionize this sector by speeding up the procedure of the arbitration process parallel to increasing the efficiency of the users. Starting in the 1950s, AI, a powered systematic structure witnessed in the era of advanced technology has spread worldwide giving people the privilege to fasten the speed of mundane work. Synching with humans, understanding their behaviour, self-correcting its omissions, and reducing deviations with the help of a reformed plan of action, AI is basically a self-learning algorithm that improves itself by its own mistakes. Numerous AI tools have been introduced to simplify the work of current-generation lawyers to draft various documents in corporate and litigation spheres and have become an inevitable part of the legal system. This brings up a debate about whether AI can replace a human arbitrator disrupting the arbitration sector and what tasks can be performed by AI in this field to assist lawyers.
The research paper aims to balance out the various opportunities in relation to AI in the Arbitration field and to analyse whether the increasing trend of using AI can also lead to the adoption of AI arbitrators in place of humans. The paper also deals with providing a thorough understanding of the challenges faced by the legal sector due to the inception of AI in arbitration which can cause hindrances to its relevance in the field and the scholar also aims in providing the best possible suggestions to mitigate those future consequences.
Research Methodology
This paper has both descriptive and analytical nature by applying the doctrinal method of research for a deep-rooted understanding of the influence of Artificial Intelligence in the field of Arbitration and its repercussions. The secondary sources used in this paper include articles, case studies, surveys, websites and journals.
Literature Review
Artificial Intelligence is considered an autonomous data system that can learn automatically from patterns or behaviours and bring changes in its features which makes the software grab the adjective “intelligent”.[1] It is true in relation to other coded data systems which need instruction to perform the task in each step contrary to the self-learning aspect of the automated data. The fact that a calculator can only perform tasks that are instructed to it by a coding system makes it incompetent to take decisions by itself as it is a pre-planned developed system that cannot adapt to changes on its own.[2] If wrong data has been coded where the calculator calculates 2+2=5, it would not be competent enough to change on its own as it is a pre-instructed system but if the same scenario is exposed to an AI then after looking into the routine patterns, it can modify its answer from the former to the correct answer i.e. 2+2=4.
This difference between machine (calculator) and machine learning (AI) can become helpful in the field of Arbitration due to its modification capacity since there is always a need for a better argument, decision, and judgment just like in the litigation field. And one of the significant reasons to introduce AI in Arbitration is the pandemic, Covid-19 which almost turned everything digital unleashing a whole new potential to bring in AI assistants for work even during a natural calamity and also to conduct various procedures of the arbitral proceedings through AI like- recruitment of arbitrators, reviewing of large documents, helping in legal research, etc.
Is there a demand for AI in the Arbitration field?
There is a demand shift happening from litigation to ADR due to delayed procedures, long waits for justice and tedious processes from the former whereas “Arbitration has been the buzzword in the commercial-legal ecosystem”[3] due to its swift and effective resolution. Since this shift was adopted to give a speedy procedure to the public, the burden to improve and provide a much more expeditious delivery of justice lies in the ADR mechanism. And this efficiency can be reached by taking the help of technology i.e., Artificial Intelligence. But because of the complex procedures of Arbitration, AI can not simply enter the market and perform any task rather it can help to boost the efficiency and up-skill the requirements of arbitrators and other bodies involved in this field.
Arbitration involves assessing large case files and facts, which require a lot of time and can be shifted to AI instead of engaging additional personnel. This can be leveraged to streamline the arbitral procedures, expedite case development, and enrich the parties’ insights and experiences. And since AI is just a self-learning tool and can just bring out changes in its outcomes, it cannot discover or create any invention or scientific innovations as it does not have a brain like humans which can bring creativity into the world. Instead, it can facilitate the creators and innovators to fasten up the time period to reach the end goal by providing assistance in research and solving complex problems. Some of the significant tasks that can be performed by an AI are listed below:
- Drafting and Analysing– In earlier days lawyers spent most of their time drafting documents but since the birth of AI tools firms and lawyers are engaging automation tools in drafting documents like Gavel (automation tool for lawyers) which help around 90 percent of the time saving from drafting lengthy documents[4]. And around 26 percent of firms with 100 or more attorneys utilize AI-based technology tools according to the American Bar Association Legal Technology Resource Centre’s ABA TECHREPORT 2019[5] which tells the incoming removal of human lawyers from drafting documents.
Also, for analysing large volumes of files, AI can be used for time-saving which can scrutinize data in seconds.
- Speedy Recruitment Procedure- AI, a miniature or a replica of a human can be dealt with even in the recruitment policy of arbitrators in the legal industry. If both the disputed parties are not in the favour of the arbitrator or have no confidencein the expertiseof the arbitrator then within a short span of the time bind with rush case handlings, this intelligent algorithm can handle the selection process by reviewing their past performances in handling cases and analysing the right attitude required for the appointment of the right arbitrator.[6] This gives the disputed parties a very transparent and appropriate justice that is uncompromised from the negligence or incompetence of an arbitrator, making the arbitration procedure more reliable and fair.
- Predictive scrutiny and judgment- In the Arbitral proceedings, there are circumstances and decisions made by the arbitrator, and often the confusion leads to clashes and chaos among the parties. With AI’s unprejudiced nature and phenomenal decision practice, strife can be neglected as the pace of arriving at the conclusion will be shortened. To forecast results or determine the chance of success for particular arguments or claims, AI systems can be trained on historical data from prior arbitration cases.[7] This can assist parties in assessing potential hazards and coming to wise judgments. Through AI’s analytical and methodical skills, the assessment of the rights and wrongs will help the arbitrator ease the decision-making process.
- AI-Language Translator- Generally, in any dispute between parties that speak a foreign language, a translator is required to untangle the language barrier that may occur between the arbitrator and the disputed parties. Due to the interpretation system, the legal proceedings will be delayed, and the system will be messed up due to the piled-up cases yet to be solved. Here, the advanced algorithm AI is capable enough to comprehend foreign wordings and be able to properly conclude the case with the right and quick momentum.[8] These translators can help in bridging the gap between language and justice in the Arbitral proceedings.
- AI Arbitrators- In 2016, the judicial decisions of the European Court of Human Rights (ECtHR) have been predicted to have 79% accuracy using artificial intelligence (AI)[9].
Giving AI the power to lead and make decisions will not only help to solve messed up cases expeditiously but also will have a hand in making precise decisions and bringing out efficacious results. The ADR system can grab the full advantage of the advanced data by eradicating the prolonged process of training being held up along with the satisfaction due to fulfilled expectations of AI that they will not reiterate the mistakes as commonly done by humans.
All these tasks can be effectively and efficiently performed by Artificial Intelligence and it also proves the existence of demand of AI in Arbitration for the same. But the question that can be raised by the last advantage mentioned above is, Can the demand for AI in the Arbitration field be superseded by a need to introduce AI as a mainstream artist instead, of a facilitator, rendering people unemployed and jobless? The answer to this is AI is just an algorithm and choosing AI above humans is a big step toward the path of destruction. As firms, lawyers or any individuals start to choose and give importance to AI as a mainstream worker in any field, for their own progress and development then not so far is the situation where the world will be dictated only by algorithms with no space for humans.
Firstly, to take the lead as a sole mainstream arbitrator, Artificial Intelligence has to compete with the Human Intellect which is the creator of the former. These self-learning algorithms cannot be said competent enough to enter the legal sphere as arbitration even though some robot lawyers like ROSS, and Joshua, powered by AI have already been hired by firms. If an AI like ChatGPT is taken into consideration then it is not that wise enough to handle manipulations as it can easily be fooled with certain statements written in an indirect way. These automated systems have a wide range of limitations which will be discussed in this paper.
Biased but Not Biased?
An AI arbitrator is considered to have unbiased opinions and judgments as it acts according to the data it has been fed but the irony is that if the data which is being fed to the system is only corrupted by an unconscious bias, then the system can never give an unbiased opinion. This is a strong catch-22 where a system that is incapable of having biases is itself providing biased decisions or judgments due to the distorted data. A system that is taught that women can only do household chores and has a patriarchal outlook can never give a decision based on deviation from this thought as the algorithm does not have the creative ability to think like a human mind.
Similarly, lacunae persist in the predictive analysis performed by the data system as it can only give predictions or judgments by analysing the historical cases even though these cases are caught by an unconscious bias towards a specific group, section, gender, etc. One more drawback is that when the self-learning systems are exposed to a type of case which needs to be looked upon from a different perspective as compared to the previous database, it will fail to give a judgment or decision beyond the scope of the available data. This calls for a human arbitrator who has the intellect to understand and analyse the cases beyond a limited scope of cases and bring in new perspectives by considering the welfare and justice of the parties.
A Threat to Data Privacy
The arbitration is all about confidentiality and privacy of the personal information of clients. As noted by The Brookings Institution’s Artificial Intelligence and Emerging Technology in its report, “As artificial intelligence evolves, it magnifies the ability to use personal information in ways that can intrude on privacy interests by raising analysis of personal information to new levels of power and speed.”[10] Data installed in AI are very complex and enormous which makes it difficult to be depended upon the security of personal information. Because of its large chunk of info, there are chances of unauthorized access and infringement of the algorithms which can misguide AI leaving it confused and unprecise. The parties will lose out on their faith in the arbitrator because of the inability of the ADR system as they set a criminal free due to the inability to secure their client’s personal sources.
Due to the breach of privacy of the personal information of clients, the data protection laws might not be very helpful as consent will be needed from those clients for the assessment of their information to retrieve those data back which can be strenuous to do so causing more hindrance than conduciveness. Hence, a better data protection ability is needed to be ingrained in AI systems.
Unemployment on the verge
One of the most daunting consequences of relying on AI is the preference of AI arbitrators over human arbitrators, AI translators over human translators and the performance of a major chunk of tasks by AI instead of a lawyer in arbitration. The fact that AI can render speedy service as compared to humans may in the coming future increase the graph of unemployment. In 2018, a report from the World Economic Forum estimated that AI would create a net total of 58 million new jobs by 2022. But will also displace or eliminate 75 million jobs by 2022[11] This can lead to a destructive situation where humans can be replaced by AI.
Another element in relation to unemployment is the reduction of humanity and empathy in people to increase their progress by using AI. One of the incidents was when an Indian CEO named Sumit Shah fired 90 percent of the employees and replaced them with an AI chatbot claiming that the latter reduced 85 percent of the customer support costs. This tells the growing expectations of efficiency and accuracy above human capacity.
It should be looked at in a way that when any creation is disrupting the very existence of its own creators that is humans then it should be kept on hold. For instance, the creation of Ride Sharing Apps like Uber and Lyft was very useful for users and was giving top-notch service for travellers to ride with ease and comfort. But because of the rise in Uber customers, the local taxi drivers whose sole breadwinner was to provide service to travellers have been losing out on their only means of earning income which led to the banning of the app.[12] Similarly, the usage and involvement of AI systems in arbitration where it can render people jobless should be put a pause, as if not today then most probably in the future the same situation like the ride-sharing apps can come where AI will have to be either banned or be reduced to the job of a facilitator again.
The fact that technology is being modified and developed for humans to elevate their efficiency, there should not come a situation where humans are facing any difficulty due to any advancement.
Suggestions and Conclusion
Technology is introduced for humans and not the other way around. The introduction of AI in the field of arbitration should be applied to the extent it doesn’t jeopardize the survival of humans. Since there is pressure, especially on developing nations to keep up with the ongoing developments in the world to avoid missing out but it has to also keep in mind the various other issues the nation has to go through like unemployment and huge capital investment in consequence of using the same. Basically, arbitration is reviewing facts and case files which takes a lot of time and can be transferred to the AI instead of employing more humans. So, AI can be infused in arbitration where the saturation level of employment has reached but the work which is mainly routine work can get completed through AIs like contract drafting, analysing large files and presenting it to humans. AIs should be implemented in arbitration in such a way that it increases the quality of the efficiency a human can produce and improves the quality of work instead of seizing the very existence of humans by replacing them as arbitrators, as this might give more accurate and fast result but later it will definitely need to be removed altogether from the systems due to affecting of the right to work of humans. In Italy, AI got banned due to questions of data privacy, and arbitration is all about confidentiality which these algorithms are prone to hacking and failures. This explains how AI can be amateurish in handling cases on its own as an arbitrator and simply relying on an algorithm instead of humans can be a wrong move. These obstacles and failures in the path of AI in Arbitration weaken the scope of a future of AI replacing humans. The same statement was used when computers came into the market that it will create unemployment but people got adapted to the change and more jobs were created. Similarly, a situation where human arbitrators will be replaced by AI arbitrators is a distant dream as it will definitely be against the human race and cannot work a long way.
References
- Brian Bushard, Will AI Take Your Job? 27% Of Jobs in Wealthy Countries at High Risk, Report Says, Forbes, Jul 11, 2023,
- Cameron F. Kerry, ‘Protecting Privacy in an AI Driven world’, (2020): Report of the The Brookings Institution’s Artificial Intelligence and Emerging Technology
- Sushree Swagatika, Impact of (Artificial Intelligence) AI in Arbitration, IP & Legal Filings, March 19, 2021
- Dr. Aline Tanielian Fadel, Predictive Analytics and Diversity in International Arbitration: Friends or Foes? , The American Review of International Arbitration, October 8, 2021
- Numan Zargar, From litigation to arbitration: a paradigm shift, Greater Kashmir, 18 Nov 2021
Sanskriti Banka
Christ University, Bangalore
[1] Kathleen Paisley & Edna Sussman, Artificial Intelligence Challenges and Opportunities for International Arbitration, 11 N.Y. DISP. RESOL. LAW. 35, 35 (2018), https://sussmanadr.com/wp-content/uploads/2018/12/artificial-intelligence-in-arbitrationNYSBA-spring-2018-Sussman.pdf
[2] See Jacob Turner, Robot Rules—Jacob Turner, LAW POD UK (Mar. 4, 2019, 10:00 a.m.), https://audioboom.com/posts/7191406-ep-71-robot-rules-jacob-turner
[3] Numan Zargar, From litigation to arbitration: a paradigm shift, Greater Kashmir, 18 Nov 2021, https://www.greaterkashmir.com/editorial-page-2/from-litigation-to-arbitration-a-paradigm-shift
[4] Study Shows Gavel Saves 90% of Time Spent on Generating Legal Documents, Gavel.io, https://www.gavel.io/resources/study-legal-automation-saves-90-percent-drafting-time
[5] ABA TECHREPORT 2019, American Bar Association Legal Technology Resource Centre, https://www.americanbar.org/groups/law_practice/publications/techreport/abatechreport2019/
[6] Sushree Swagatika, Impact of (Artificial Intelligence) AI in Arbitration, IP & Legal Filings, March 19, 2021, https://www.ipandlegalfilings.com/impact-of-artificial-intelligence-ai-in-arbitration/
[7] See Dr. Aline Tanielian Fadel, Predictive Analytics and Diversity in International Arbitration: Friends or Foes? , The American Review of International Arbitration, October 8, 2021, https://aria.law.columbia.edu/predictive-analytics-and-diversity-in-international-arbitration-friends-or-foes/
[8] Ibid.
[9] UCL, ‘AI predicts outcomes of human rights trials’, (October 24, 2016), https://www.ucl.ac.uk/news/2016/oct/ai-predicts-outcomes-human-rights-trials
[10] Cameron F. Kerry, ‘Protecting Privacy in an AI Driven world’, (2020): Report of the The Brookings Institution’s Artificial Intelligence and Emerging Technology
[11] Brian Bushard, Will AI Take Your Job? 27% Of Jobs in Wealthy Countries at High Risk, Report Says, Forbes, Jul 11, 2023, https://www.forbes.com/sites/brianbushard/2023/07/11/will-ai-take-your-job-27-of-jobs-in-wealthy-countries-at-high-risk-report-says/?sh=4f5858591113
[12] Will Sowards, Which Cities Have Banned Ride-Sharing Apps? Passport Health, June 13, 2018, https://www.passporthealthusa.com/2018/06/which-cities-have-banned-ride-sharing-apps/