Masters Information Technology Law Dissertation Sample
Here is a sample that showcases why we are one of the world’s leading academic writing firms. This assignment was created by one of our expert academic writers and demonstrated the highest academic quality. Place your order today to achieve academic greatness.
Understanding the Role of Artificial Intelligence and its Application in Online Dispute Resolution.
Artificial intelligence comprises features like solving the issue, developing rational judgment, and learning that benefit the legal domain. Settlement of disputes through online means consumes less time and money and satisfies the disputants. Using AI-based systems for online dispute settlement leads to more effective ways of conflict resolution.
This study analyzes the prevailing online dispute settlement methods, the kind of available resolutions, AI techniques to make conflict resolution more efficient, and its application in the law domain. It also evaluates the advantages and disadvantages of online issue settlement and evaluates the challenges and risks associated with adopting artificial intelligence in online dispute settlement.
Artificial intelligence, Online dispute resolution, Artificial intelligence in law.
There is an extensive history that how humans happened to resolve disputes (Zeleznikow, 2021). Recently, researchers have been using artificial intelligence to examine and resolve disputes in an improved and optimized way. At the initial stage, the Online Dispute Resolution (ODR) method was mainly used to provide information to parties and to contact them (Lodder and Zeleznikow, 2010). However, currently Artificial Intelligence (AI) based technologies are applied for ODR processes for planning, devising strategies, and decision-making. ODR systems application is not only confined to e-commerce disputes but is also widely applied in the legal domain area. After the development of e-commerce during the mid-1990s, the scope of using technology techniques for dispute resolution has also grown (Schoop et al. 2003). The increase in issues, regardless of the location indicates that the disputes need to be resolved online. Thus in this research, the focus is to analyze those intelligence procedures useful for human negotiators.
Latifah et al. (2019) explain that ODR has become paramount important with e-commerce development. This method of dispute resolution takes place entirely or partially online and is based on using the ICT (information and communication technology) method. The dispute resolution by using the ODR technique is not only confined to cyberspace rather it goes beyond it as well (Omoola and Oseni, 2016). Alternate Dispute Resolution (ADR) procedures resolve the issues (Latifah et al. 2019). As per the study by Schmitz and Wing (2021) under ODR, the resolution order is processed with the involvement of four parties such as a claimant or the party who initiates, then comes the respondent, at third a neutral party is also incorporated. Lastly, the fourth party comprises a technology-based intermediary. Bello (2018) believes that a fifth party is also involved in the process, which is the individuals providing service associated with the technological component. Carneiro et al. (2014) discuss the two classifications of ODR as per their level of autonomy which is called first-generation and second generation. According to the study, the human component is the key player in deciding disputes under first-generation whereas the intelligent system is not a part of it.
Moreover, the role of electronic tools is to provide facilitation to the involved parties to communicate and coordinate. The key technological tools utilized are quick messaging facility, audio, video calls and video conference, etc. The second generation mainly comprises technology; intelligent agents plan, devise strategies, and make decisions. Techniques of AI as well as Mathematics and philosophy are deployed to make the agent act like a human. However, there are certain advantages and disadvantages of ODR also. One of the most accepted definitions of AI is the capability of a technological tool such as a computer or robot or a software program to carry out tasks through the use of human intelligence (Pannu, 2015). It explains that the key features associated with intelligence that is analyzed in AI are cybernetic adaptation and learning, theme creation, reasoning, solution of problems, visual recognition and examining, language features such as understanding of speech, information processing, and restoration, etc. Gadde and Gadde (2020)  determine the high interconnectedness of AI in the field of areas such as robotics, modeling, and video games.
Thus the study believes that AI is not only confined to the computer sciences field rather it is an interdisciplinary field that is inclusive of biology, linguistics, engineering as well as psychology, and mathematics. With the exponential growth in the use and performance of computers, the scope of AI use is getting wider day by day in fields of human lives which are well-known subfields of computer (Tien, 2017) . Virtual programs such as Google and Siri, vehicles and drones, weather forecasting, recommendation systems like Netflix and Spotify are only a few examples of AI usage in our daily life. The study analysis that apart from the field of computer AI techniques is also used in medicine, engineering, economy-related work and finance, ecology, business, and law as well. The reason for this wider application is for AI to be less costly, quicker, and faster and provides consistent and long-term output as compared to natural intelligence. The study by Zeleznikow (2021) examines that during the period of the last six years AI is rapidly developed to be used in ODR systems and these solutions are easily accessible in the market.
The study analyzed that the main focus of the development of AI for ODR has been remained in creating user-friendly designs because the system provides aid to everyone even those who are not professionals. The study asserts that user designs have got even significantly important during the prevalence of the Covid-19 spread. The pandemic hit at the global level has increased the reliance on the use of AI for ODR even more and it requires ODR to present facilities more than sheer video conferencing facility. This research provides a conceptual overview of the role of AI-based system for ODR including law and justice context through explanation of legal procedures and forms and how to utilize expert knowledge for ODR. By using simpler technology it covers how AI works in the domain of law and justice.
Objective of the study
Following are the research objectives:
- To critically examine the role of AI application in the resolution of online disputes and the challenges involved in the process.
- To measure the usefulness of AI in online dispute resolution for parties.
- Examining the AI in online dispute resolution for systems such as legal agreements and supporting communications.
There is one research question:
- How AI is useful for effective online dispute resolution and what are the challenges which hinder AI role to be fully utilized in online dispute resolution?
The chapters of the study are organized as chapter 1 is about the introduction of AI and ODR and their application; Chapter 2 presents systematic literature on the topic. Next is Chapter 3 which provides research methodology; subsequent data analysis is detailed in Chapter 4; lastly Chapter 5 presents the conclusion of the study.
 Zeleznikow, J., 2021. Using Artificial Intelligence to provide Intelligent Dispute Resolution Support.Group Decision and Negotiation, pp.1-24.
 Lodder AR, Zeleznikow J., 2010. Enhanced dispute resolution through the use of information technology. Cambridge University Press.
 Schoop M, Jertila A, List T., 2003. Negoisst: a negotiation support system for electronic business-to-business negotiations in e-commerce. Data Knowl Eng 47(3):371–401.
 Latifah, E., Bajrektarevic, A.H. and Imanullah, M.N., 2019. Digital Justice in Online Dispute Resolution: The Shifting from Traditional to the New Generation Dispute Resolution. Brawijaya Law Journal,6(1), pp.27-37.
 Omoola, S.O. and Oseni, U.A., 2016. Towards an effective legal framework for online dispute resolution in e-commerce transactions: trends, traditions and transitions.IIUM Law Journal,24(1).
 Schmitz, A.J. and Wing, L., 2021. Beneficial and ethical ODR for family issues.Family Court Review,59(2), pp.250-267.
 Bello, A.T., 2018. Online Dispute Resolution Algorithm: The Artificial Intelligence Model as a Pinnacle.Arbitration: The International Journal of Arbitration, Mediation and Dispute Management,84(2).
 Carneiro, D., Novais, P., Andrade, F., Zeleznikow, J. and Neves, J., 2014. Online dispute resolution: an artificial intelligence perspective.Artificial Intelligence Review,41(2), pp.211-240.
 Pannu, A., 2015. Artificial intelligence and its application in different areas.Artificial Intelligence,4(10), pp.79-84.
 Gadde, H. and Gadde, U., 2020. Artificial Intelligence and its Applications.Artificial Intelligence,9(7).
 Tien, J.M., 2017. Internet of things, real-time decision making, and artificial intelligence.Annals of Data Science,4(2), pp.149-178.
This chapter discusses the previous work on the AI applications in ODR from different perspectives. Moreover, it explains the advantages and disadvantages of ODR, ODR procedures, and their types. The chapter also analyzes the use of AI in ODR in the legal domain.
The use of ODR is becoming wider and it is used for litigation or in other types of dispute resolution (Goodman, 2002). One of the main advantages of the use of ODR is price and speed. Because mostly litigation is a time taking process, on other hand ODR techniques are very effective in saving time for those parties who adopt it. Moreover, ODR is also economical in comparison with litigation since it involves no charges on behalf of the court fee. The requirement of travel for the third party also obliterates which reduces the expenses because the implementation of ODR is independent of location (Kaufmann-Kohler and Schultz, 2004).
The study analyzes that the parties involved have an option of using the online platform for the solution of issues from their home which proves to be easy and affordable for them. Whether parties belong to different geographic regions, the location independence of this system makes it handy and efficient for the resolution of the problem. The ODR system has a mechanism for parties’ coordination and providing them right for partly controlling the outcomes of the process (Mania, 2015). Thus the system bounds the involved parties to collaborate during the process to obtain resolution or settlement of the dispute and to control the outcome. Hence the ODR system enables the parties to reach a solution and agree on it without the court’s legal imposing.
The study views that it appears less intimidating for the involved parties to deal through the online environment as compared to stand before a judge. Therefore, the parties usually feel no limitations on exchanging opinions. Moreover, they pay attention more to the facts because they do not personally see their opponents; hence less emotion is involved in the deal. In this way, communication goes harmonized and provides the parties time to find the answer which is more objective and involves fewer emotions. Further, the storage of formal documents and correspondence enables it to be used later for evidence purposes.
The study by Qutieshat (2017) analysis the limitations of using ODR. Though ODR is very effective in some issues, however, in many cases it is preferred by parties to contact personally. Particularly when the agreement comprises a lot of details or as an example, the dispute is about a family matter then the parties feel emotions involved which also need due consideration. ODR also highly depends on the availability of tools such as internet connection and convenience to work on a computer and availability of the relevant platform. Furthermore, Wiszniewski (2015) believes that regardless of numerous advantages of using ODR, when there exists no physical contact among the parties it can lead to potential issues. The study explains that the parties’ personal contact develops personal relationships hence it results in using different language and tone which is observed absent during ODR. Moreover, the higher authority body’s absence serves as a loophole for the party which is intended to cheat (Ebner and Zeleznikow, 2016). Also if tools of seeing each other such as video conference are not available then it’s not possible for both parties to analyze body language or there is a likelihood of parties’ discomfort to face the camera while speaking.
However, the study gives the solution of telepresence which creates a physical presence for disputing parties and mediator. The study also argues that telepresence and videoconference cannot provide a substitute for one on one communication which develops trust and support for each other. The study also points out another issue which is security, since the ODR processes involve exchange of many confidential information which also comprises secret trade information. The use of cyber technology is also exposed to cybercrime and hence any sensitive information leakage can cause damage to both parties (Van Arsdale, 2015). The study argues that for ODR to be effective and efficient it requires many resources. Therefore, firms are not willing to let others access the system freely and they patent a particular procedure that obstacles the ODR spread.
Process of ODR
The study by (Condlin, 2016) discusses the ODR process; this process begins with a party’s request who tends to resolve a new issue. The party also provides necessary information relating to the other participants who are then approached and their willingness is sought to coordinate. When the involved parties’ mutual understanding is sought, then the system collects issue-related details and information. Example of such information is monetary values, the involved parties expectations, emotions and evidence-based on documents. Because it also includes feelings and emotions, therefore, autonomous ODR providers find this stage the hardest. Subsequent stage analysis all the gathered information which produces outcomes using the system which is based on human or technology. The final stage comprises presenting outcomes to the parties.
The Best Alternate to a Negotiated Agreement (BATNA) is a key and basic mode used in a principled negotiation that concentrates on aims criteria and interest. It does not focus on disputes related to different people and their position (Wilson-Evered and Zeleznikow, 2021). The study explains that the purpose of negotiation is to generate improved results which otherwise are not possible to obtain. If one has no awareness of the outcomes that can be acquired from unsuccessful negotiation, he/she undergoes the risk of either engaging in an agreement which will turn him better off than rejecting it or he will get the understanding of rejecting an agreement which would make him better off than entering into it. The study finds that ODR process takes BATNA as standard, therefore, ODR process’s importance is widely understood. The study believes that the ODR process provides solutions that are taken better than BATNA.
Procedures in ODR
ODR composition contains 74% of mediation and 40% consists of negotiation and providing advice to the clients (Carneiro et al. 2014) . Among different dispute resolution techniques, the degree of control by the parties also differs. Seeking dispute resolution through the involvement of the court provides the parties near no control over the results. Whereas in eNegotiation the parties get complete control of the process for the result to be accepted or rejected (Zlatanska and Betancourt, 2013) . The study argues that in different techniques of dispute resolution such as facilitation, arbitration and negotiation the degree of control is maximum by the parties. The study by (Thompson, 2015)  explains that the involved parties are advised to initiate negotiation. If it becomes necessary then the next part of negotiation also involves mediation and finally they engage with the arbitration. However, if the issue is still not solved, then the failure is communicated and the alternate solution is sought. The study explains the procedure as per Figure 1.1.
Figure 1.1: The solution paths.
eNegotiaion or the process of Negotiation is an informal procedure that involves the collaboration of the parties to resolve the issue by themselves (Ebner, 2021). There exist a few classifications of negotiation; however, two types of negotiation automated and assisted are explained by the studies. The automated negotiation comprises a blind bidding procedure and this technique is more appropriate for monetary issues such as the issues in the agreement of buying and sell and divorce having no children. The settlement is maintained if the bids stay in the agreed-upon range by the parties which are commonly 20% to 30% (Simkova and Smutny, 2019). The agreed range of the bids enables the computer to compute the median amount and the issue is then settled otherwise the involved parties bit for another time. The study explains the other form of negotiation i.e. an assisted negotiation which is more sophisticated. The involved parties are provided with proper guidelines, advice, and standard type of forms using a web-based platform of communication by the ODR providers. It gives thorough guidance to the parties on proceeding to solve their issues.
On the other hand Bales (2017) explains distributive and integrative approaches relating to negotiation. The distributive negotiation sees the issue as something which can be distributed and divided among the parties with the motive to get their maximum satisfaction. Under integrative negotiation, the issue is believed to have solutions more than ones that appear at first sight. Shepherd and Landry (2013) demonstrate it through the example of oranges to exemplify the difference. For example, two females require orange but the fruiter contains only one orange. Using the distributive approach, the females make the remaining last orange in two equal parts and each of the women get an equal share which is one-half. However, both females get 50% of what they wish to get. Under the integrative approach, both the females discuss what both they wish to have and get to the point where they both feel satisfied. In such a case one female gets the orange and the other gets the peel of the orange for tea. Thus applying this approach both the females receive 100% as per their wish.
Under this negotiation procedure both the parties choose a mediator who is independent and unbiased. The mediator facilitates the process through communication and helps both parties to settle the issue by themselves (Tuleutayevna, 2016). The approach requires the involved parties’ willingness to cooperate. The result is not decided by the mediators; however, their role is still important because the success or failure of the issue solution depends on the competency of the mediators. Mostly the result is non-binding, however, it drives to a stable and binding solution. Usually, mediation is a successful process as the parties are guided to an agreeable settlement among themselves. Lodder and Zelznikow (2005) argue that online mediation is similar to that of offline mediation apart from using the online facility. Meetings are carried out through a virtual method using a chat room and it is conducted along with each party exclusively or parties are joined together in a meeting. Virtual mediation and electronic negotiation replace each other sometimes. The study identifies the technologies adopted by the parties like video conferencing, telephone, email, and discussion platforms.
Sela (2018) defines the procedure as more informal and flexible as compared to the proceedings in the court. In this method, a third unbiased party serves as an arbitrator who hears each of the involved parties. The arbitrator is presented with the evidence of the issue and also with the authority to decide on the issue. Opposite to meditation, the arbitrators are non-aligned and they cannot influence the involved parties’ decision, such decision tends to be binding or it can be unbinding.
Application of AI in ODR
The potential role of AI systems to provide increased access to justice is recognized for justice and ODR (Lodder and Zelznikow, 2005). AI generates synthetic intelligence by using the source of technology, therefore, some work in this field is attempted to generate tools that demonstrate the human thought pattern (Sharmila, 2020). The study by Kravec (2006) also supported the argument that systems are struggled to create to perform tasks or produce output based on human intelligence. The outcomes of the output may replicate human intelligence regardless of the reasoning process dissimilarity with human thought. The ultimate demonstration of the AI system is to create tools that can cast a deep effect on human thinking or logic processes. AI systems’ classification and uniqueness are based on reasoning techniques. AI may be based on deductive reasoning, determined based on the case or it can be hybrid (O’callaghan, 2003) . AI is a very wide area and its topic ranges from a calculator used in a phone to independent systems. The study by Keplesky (2007)  discusses AI’s three levels as per its quality:
ANI is artificial narrow intelligence or is also called weak AI which is specialized to perform only a single task such as a fine chess playing by a machine; however, it is unable to perform anything else because the system is not capable of thinking. Shendre et al. (2018)  provide examples of this type of AI such as Samsung produces washing machines that are based on the pre-programmed pattern. Apple has Siri and spam filter for email and Google translator etc.
AGI is artificial general intelligence which is also called strong AI (Keplesky, 2007). These AI-based systems either equalize human intelligence or their performance can exceed it. AGI is capable of doing any intelligence-based task which is also performed by a human. The study believes that currently no AGI system is recognized, however, the study provides such machines examples in The Terminator and Iron Man movies.
Keplesky (2007) defines ASI also called Artificial Superintelligence as an intelligence that is very smarter than the finest human intelligence in every area practically which also includes scientific creativity, social competencies, and general intelligence. However, no such system of ASI is known so far. Currently, ANI is created which enables computers to think for performing logic tasks, to translate languages and or calculus, however, to perform actions like movement and vision it has not succeeded (Moravec, 1998) .
AI in Justice System
Since 1970 efforts were made to use AI in the legal system (Lodder and Zelznikow, 2005). The study finds that most of the efforts were made in terms of capabilities and complications. Zeng et al. (2020)  also support the argument that initial attempts were made to create judges who are computerized and can carry out legal reasoning of complicated nature. Besides this other efforts were made to get a fine interpretation and analysis of rules under law as well as to get the help for advanced decision making in legal matters. Meldman (1977)  also argues that it was also attempted to get analysis of law pertinent to a particular rule and also picking rules from particular cases and formulating reasoning based on cases.
However, Zeng et al. (2020) find that over time, the broader application of such sophisticated programs proved difficult to obtain. The reasons for failure were the high cost associated to create such specialized programs and there was limited demand for them. Till the 1980s the unavailability of up-to-date internet also made it difficult to reach out to such systems and keeping them updated. At that time the roles for such systems were also not well defined which hindered the broader application of it. However, during the 1980s the notion was developed that AI in the legal system can prove a supportive tool and something near to human performers. Debessonet and Cross (1986)  find that the initiative takers suggested including jurisprudential theory to be the part of designs. However, in recent years AI systems are considered as independent systems and practical tools to be engaged in legal proceedings using the advanced foundation.
However, the study by Thompson (2015) finds that regardless of failures, in the beginning, the work on AI initiatives was continued with ambition and it was attempted to develop web advisers on legal matters. The study explains that legal web advisers depend on AI to present legal advice via computers. The systems are developed to gather data and facts from non-lawyers using the method of interviews techniques and generate answers by employing a decision tree analysis. JPES (Justice pathway expert system) is based upon the same model, however, the legalistic focus observed in this system is lower. Under JPES legal guidance can be the part of the output that relies on particular circumstances for the non-lawyers and generates it as a component of larger output based on guidance.
The study by Davenport and Short (1990)  explains that a computerized system that uses a comparatively easy and simple AI can be deployed for the majority of the population to serve in legal matters effectively and on time. The stud argues that the business world is adopting technology to facilitate firms to get free from past practices which no more serve the functional purpose. Thus JPES has the tendency to facilitate the justice system to break away from past obsolete practices and enter into the present (Tyler, 1987) . However, the traditional stakeholder in the justice system believes that users’ satisfaction depends upon the result obtained through the issues resolution process. The study argues that as per psychological research people give substantial weight to their response to a procedure.
Thus Thompson (2015) suggests that JPES should be formulated based on being functional for the ones who got the negative impact of their emotional situation. The study suggests that regardless of the failure of broader application and use of AI in the legal field, further exploration for its application should be made. Gélinas et al. (2015)  find that when it arises a crisis scale for the issue in reaching out to justice then developments continue for technology. Hence JPES is aimed to be a practical, suitable and realistic process to be applied and used on a broader level. Once it succeeds, it could be capable of being helpful to develop the AI technologies related to justice and also could provide improved access to ODR procedures as well as justice. The development of AI in the legal field will be helpful in the development of AI in general. Makridakis (2017)  analysis that the progress in building machines for completion of particular works in ways that were never possible before gave birth to a new dream for AI. For justice purposes, JPES is continuously shifting to more easy and practical output which is also less sophisticated thus it promotes more adopting of such systems.
The study by Williams et al. (2013)  examines that the use of technology in ODR is criticized for computer-based machine’s inability to understand the involved parties’ feelings and their social requirements. The process of human interaction with computers gives the basis that how humans can utilize and interact with technology. Thus it poses a challenge to this criticism. Meany and Clark (2010)  support the argument by finding that the interaction of humans and computers is similar to the method of interaction which is from human to human. Parties understand the fact that their interaction is with the technology-based tool, however, they are inclined to go by the social rule of interaction between human to human and they behave with machines like they are also with emotions and feelings.
The study finds that human tends to get angry towards the technology which is computer-based or they can feel appreciated or flattered. Users are aware that computer is unable to feel emotions or develop thinking but still they follow a social method without conscious efforts as similar to humans mutual interaction. These behaviors are alike from initial textual interfaces to advance communication levels. Dourish and Bell (2011)  find that undergoing a conflicting scenario results in cognitive changes due to changes in emotions which has an influence on our thinking and way to act or respond. The intensity of the conflict rises with the increase in tension among the parties. When such a conflicting situation arises it can cause a concentration loss in parties. Thompson (2015) reveals that people’s access to justice processes and machines via based on the internet often come across emotional distress.
AI Technologies Applied in ODR
AI technologies enable ODR procedures to handle the issues like complicated multiparty, multiple problems, and multiple contracts (Carneiro et al. 2014). AI technologies are used to get help in decision support systems, expert systems, knowledge-based systems, and rule-based systems, etc. The study examines that these approaches are sometimes used together such as decision support and multi-agent systems. AI techniques are applied in the area of the resolution of problems. These solutions can be formulated based on past cases or Neural, Bayesian processes or ontology, genetic algorithms methods. The study discusses the following systems:
DSS or Decision support system comprises systems that combine necessary data from raw to complicated systems to obtain the fairest results (Wilson-Evered and Zeleznikow, 2021). The areas like banks, insurance firms, and social security-related institutions use these kinds of systems. The study analyzed that in-law area the DSS helps the users to investigate huge data and information which is useful to get improved results. DSS tools usually comprise rule as well as case-based reasoning, neural system, and machine learning. Spilt-up is taken as a common example of DSS.
Expert system is explained by Zeleznikow (2016)  as computer programing which are designed and developed by human experts and enable them to perform equal to or sometimes more than human experts in that particular field. Surden (2018)  analysis the competency of the expert system such as transparency, flexibility, and heuristic characteristics. The expert system takes guidance from the previous cases or the human expert supervisor adjusts its learning path as per input and output which is expected and another output which is verified. Both DSS and expert system differ each other since DSS help the human expert in decision making and providing counsel whereas expert system in the legal field perform like human experts and provide advice as well as make decisions. The applications of expert systems in various areas such as medicine, finance, and procedures control and in others are several. However, due to huge data and information in the legal field and a higher rate of disputes practitioners find it challenging to process information in the absence of any support. Thus expert systems provide help to legal practitioners to handle the cases and also provide guidance to them keeping in view the previous cases and rules. Moreover, the study reveals that simple tasks are improved through this system and lead to more effective and quick performance.
KBS or Knowledge-based system is explained by Mustajoki (2007)  as the system is based upon knowledge and KR (knowledge representation) has a fundamental role in this system. KR symbolizes the formal way of thinking since in the legal field knowledge is complex. KR takes methods from more domains and rules as well as a formal form of inference is based upon logic. Moreover, ontology provides definition of the things in the field and helps in the calculation of the procedure (Gómez-Pérez et al. 2006) . KBS is useful to provide the definition of a model which is capacitated to handle a group of data or information that is heterogeneous such as facts, logic, arguments, involved parties’ information, rules, and previous related cases. Mustajoki (2007) analysis that there are advantages of using KBS since it works on the idea of storing data digitally which can be accessed quickly and in an effective manner.
The logic argument can be analyzed for its validity and interpretation. The rules and judgment established through KBS work under uncertainty and based on knowledge new information and outcomes can also be generated. The types of knowledge are differently used in the system such as defining vocabulary, heuristics, describing attitude, rules, beliefs and norms, procedures, assumptions, hurdles, general knowledge, etc. (Carneiro et al. 2014). The outcomes obtained from new cases help in improving the existing knowledge. The study analyzes that a system to extract knowledge and storing it to a base and then an inference engine makes this whole process. Thus knowledge acquisition and implications are the fundamentals of the KBS.
Intelligent interfaces are another system that provides various types of abstraction for the practitioners in the legal domain (Lodder and Thiessen, 2003) . If developers know the procedure used for solving the legal issues intelligent interface can be developed to demonstrate and represent the area of knowledge other than the form of data. In this system, users concentrate on the content and do not focus on translating the concepts and their storage in the information system of the legal domain. It works based on information relating to the legal field. Legal KR-related frameworks and practitioners’ personal preferences. The study analyzes that this system has to be capable, flexible, and advanced to interpret its action and perform as per the requirements
 Ebner, N., 2021. The Human Touch in ODR: Trust, Empathy and Social Intuition in Online Negotiation and Mediation.
 Simkova, N. and Smutny, Z., 2019. Comparison of unassisted and smart assisted negotiation in B2B relationships from the perspective of generation Y.Information,10(8), p.263.
 Bales, R.F., 2017.Social interaction systems: Theory and measurement. Routledge.
 Shepherd, T. and Landry, N., 2013. Technology design and power: Freedom and control in communication networks.International Journal of Media & Cultural Politics,9(3), pp.259-275.
 Tuleutayevna, K.V., 2016. Alternative dispute resolution in the USA: analysis of main advantages and disadvantages.Вестник Института законодательства и правовой информации Республики Казахстан, (1 (42)).
 Lodder, A.R. and Zelznikow, J., 2005. Developing an online dispute resolution environment: Dialogue tools and negotiation support systems in a three-step model.Harv. Negot. L. Rev.,10, p.287.
 Sela, A., 2018. Can Computers Be Fair: How Automated and Human-Powered Online Dispute Resolution Affect Procedural Justice in Mediation and Arbitration.Ohio St. J. on Disp. Resol.,33, p.91.
 Sharmila, D., 2020. Online Dispute Resolution and Computational Law a Study on Digital Justice.International Journal of Advanced Research in Engineering and Technology (IJARET),11(9).
 Kravec, N.G., 2006. Dogmas of Online Dispute Resolution.U. Tol. L. Rev.,38, p.125.
 O’callaghan, T.A., Popple, J. and McCreath, E., 2003, June. Shyster-mycin: a hybrid legal expert system. InProceedings of the 9th international conference on Artificial intelligence and law(pp. 103-104).
 Keplesky, B., 2007. The Imago Effect: Avatar Psychology.Behind the Door SXSW07.
 Shendre, G., Tiwari, R. and Dhomne, J., 2018. Effect or Impact of Artificial Intelligence (AI) and Robotics.International Journal of Electronics, Communication and Soft Computing Science & Engineering (IJECSCSE), pp.45-48.
 Moravec, H., 1998. When will computer hardware match the human brain.Journal of evolution and technology,1(1), p.10.
 Zeng, J., Chan, C.H. and Schäfer, M.S., 2020. Contested Chinese dreams of AI? Public discourse about artificial intelligence on WeChat and people’s daily online.Information, Communication & Society, pp.1-22.
 Meldman, J.A., 1977. A structural model for computer-aided legal analysis.Rutgers J. Computers & L.,6, p.27.
 Debessonet, C.G. and Cross, G.R., 1986. An Artificial Intelligence Application in the Law: CCLIPS, A Computer Program that Processes Legal Information.High Tech. LJ,1, p.329.
 Davenport, T.H. and Short, J.E., 1990. The new industrial engineering: information technology and business process redesign.
 Tyler, T.R., 1987. The psychology of disputant concerns in mediation.Negotiation Journal,3(4), pp.367-374.
 Gélinas, F., Camion, C., Bates, K., Anstis, S., Piché, C., Khan, M. and Grant, E., 2015.Foundations of civil justice. Cham: Springer International Publishing.
 Makridakis, S., 2017. The forthcoming Artificial Intelligence (AI) revolution: Its impact on society and firms.Futures,90, pp.46-60.
 Williams, D., Alam, M.A.U., Ahamed, S.I. and Chu, W., 2013, July. Considerations in designing human-computer interfaces for elderly people. In2013 13th International Conference on Quality Software(pp. 372-377). IEEE.
 Meany, M.M. and Clark, T., 2010. Humour Theory and Conversational Agents: An Application in the Development of Computer-based Agents.International Journal of the Humanities,8(5).
 Dourish, P. and Bell, G., 2011.Divining a digital future: Mess and mythology in ubiquitous computing. Mit Press.
 Zeleznikow, J., 2016. Can artificial intelligence and online dispute resolution enhance efficiency and effectiveness in courts. InIJCA(Vol. 8, p. 30).
 Surden, H., 2018. Artificial intelligence and law: An overview.Ga. St. UL Rev.,35, p.1305.
 Mustajoki, J., 2007.Interactive multi-criteria decision support: new tools and processes for practical applications. Helsinki University of Technology.
 Gómez-Pérez, A., Fernández-López, M. and Corcho, O., 2006.Ontological Engineering: with examples from the areas of Knowledge Management, e-Commerce and the Semantic Web. Springer Science & Business Media.
 Lodder, A. and Thiessen, E., 2003. The role of artificial intelligence in online dispute resolution. InWorkshop on online dispute resolution at the international conference on artificial intelligence and law, Edinburgh, UK.
 Goodman, J.W., 2002. The pros and cons of online dispute resolution: an assessment of cyber-mediation websites.Duke L. & Tech. Rev.,2, p.1.
 Kaufmann-Kohler, G. and Schultz, T., 2004.Online dispute resolution: challenges for contemporary justice. Kluwer Law International BV.
 Mania, K., 2015. Online dispute resolution: The future of justice.International Comparative Jurisprudence,1(1), pp.76-86.
 Qutieshat, E., 2017. Online Dispute Resolution.British Journal of Humanities and Social Sciences,18(2), pp.10-20.
 Wiszniewski, B., 2015. Acceptance Testing of Software Products for Cloud-Based Online Delivery.TASK Quarterly,19, pp.495-526.
 Ebner, N. and Zeleznikow, J., 2016. No sheriff in town: governance for online dispute resolution.Negotiation Journal,32(4), pp.297-323.
 Van Arsdale, S., 2015. User protection in online dispute resolution.Harv. Negot. L. Rev.,21, p.107.
 Condlin, R.J., 2016. Online dispute resolution: stinky, repugnant, or drab.Cardozo J. Conflict Resol.,18, p.717.
 Wilson-Evered, E. and Zeleznikow, J., 2021.Online Family Dispute Resolution: Evidence for Creating the Ideal People and Technology Interface(Vol. 45). Springer Nature.
 Carneiro, D., Novais, P. and Neves, J., 2014. Conflict resolution and its context.Cham: Springer.
 Zlatanska, E. and Betancourt, J.C., 2013. Online Dispute Resolution (ODR): What is it, and is it the Way Forward?.Arbitration: The International Journal of Arbitration, Mediation and Dispute Management,79(3).
 Thompson, D., 2015. Creating new pathways to justice using simple artificial intelligence and online dispute resolution.IJODR,2, p.4.
The study by Lodder and Zeleznikow (2005) designed a three-step framework for ODR, established on AI use for effectively finding the solutions to the issues. The ODR environment provided by their modeling is a virtual space that gives a range of resolution tools to parties for their problems. However, Thiessen and Zeleznikow (2004) find that the ODR system comes across five key challenges to perform effectively; representation of the issue, preferences elicitation, the usefulness of interaction, provision of rationality, and automation scale. The present research is aimed to analyze the role of AI application in ODR, AI usefulness in ODR, and challenges involved in the process. In this chapter, research methodology is highlighted, which is used to explore the role of AI in ODR.
Research methodology presents the way a researcher systematically develops the research procedure to ensure true and real outcomes. These outcomes address the objectives and aim of the research. The study by Kumar (2018) discusses that research methodology comprises method or processes that classify, scrutinize, proceeds and choose information relating to a particular topic or problem. It enables the researchers and readers to examine the reliability and validity of the research overall critically. Allen (2017) describes that as a part of research methodology, the research method deals with research tools and procedures to conduct the research work. Moreover, the study calls the research method schemes and techniques applied in research and mainly based on scientific planning. To achieve robustness in the analysis, the selection of a relevant research methodology and research method is the key element.
The study by Zhu et al. (2018) analysis that a systematic literature review selects, classifies, and critically examines the research for the purpose to formulate research questions and objectives. The present study follows the systematic literature review methodology because it gives a transparent and comprehensive search acquired from multiple research engines and literature, which can be reproduced and reflected (Kumar, 2018). Zhu et al. (2018) argue that the systematic literature review consists of formulating a well-thought search strategy that concentrates on the relevant objectives or pre-defined research questions. It highlights the kind of information critique, searches, and finds in a known time. Snyder (2019) explains that the term search is related to search strategies that are inclusive of database networks, names, and search dates. It is confined to all the important resources to be part of the review.
As per Allen (2017), the systematic literature review deals with the qualitative research approach; this type of research handles the non-numeric and unstructured data to understand the concepts, viewpoints, and experiences. The usual tools to gather data under this research are text analysis, visual observations, interviews, and focus groups. The study believes that the qualitative research method results in enriched research with an adaptive environment. The qualitative research approach applies various qualitative research methods to address the research questions through ‘how’ and ‘why’ (Snyder, 2019). The information acquired using this approach incorporates human experiences, processing on an open-ended problem, and above, all it is cost-effective. The present research examines the role of AI application in ODR and the challenges involved in the process, through evaluating the previous researchers’ work experience and analysis.
The present research is based on the review of documents, and for this purpose, it has been referred to a literature plethora. Some authentic websites such as Google Scholar, Research Gate, Science Directory, and Academia have been utilized to retrieve the documents. Some keywords such as artificial intelligence, online dispute resolution, systems design, artificial intelligence in the legal domain, access to justice are used to reach out to the relevant research articles, papers, and books. To get more relevant and closer articles, snowball sampling has been applied in research. Parker et al. (2019) explain that snowball sampling is used to cite or give reference to the articles, selected for the study from the given literature to highlight and find the additional literature.
The inclusion and exclusion criteria have been used for snowball sampling. It has chosen only those relevant material and papers, published after 2003 except referring to some articles of 1977 and onwards when analyzing the historical perspective on the topic. Articles after 2003 are preferred because the phenomena of AI use in ODR is emerged and flourished during this period. For exclusion, the present research has taken articles only in the English language. Articles published in other languages and of different duration are excluded from the study. Similarly, websites and materials which are unauthentic such as Course Hero, UK Essay, and Wikipedia, are also excluded.
Inclusion and Exclusion Criteria
Hadi and Closs (2016) define inclusion criteria in the systematic literature review that is particular for the characteristics and features of the research. On the other hand, exclusion criteria require a certain set of elements to be filtered out and excluded from the study. These criteria improve the study’s accuracy and are effective in obtaining good results. Following is the inclusion-exclusion criteria of the study:
|· The articles published from 2003-2021
· Articles having a good impact factor and match with the keyword
· Relate to research aim and objectives.
· Supportive Qualitative and Quantitative materials
· English Language
· Complete research with abstract, practical application, and limitations.
|· Irrelevant, unauthentic, or zero cited
· Materials older than 2003
· Comprise irrelevant factors
· Complete quantitative studies
· Sources, i.e., UK Essay, Wikipedia, etc.
· Languages other than English
· Incomplete research materials
Table 1.1: Inclusion and Exclusion Criteria
Data Analysis and Synthesis
Allen (2017) describes data analysis as the most critical phase of the systematic literature review because it comprises the gathered data and information’s summarizing and its interpretation using the logical approach and analytical reasoning to find the relationship or trends. The literature in the present research focuses on finding the role of AI application in ODR, AI usefulness, and the challenges involved in the process. In systematic literature, the data analysis is helpful for the researcher to obliterate the error level because more than one author works on providing the same information with accuracy and eliminate irrelevant information or duplication (Snyder, 2019).
Kumar (2018) asserts on acknowledging the authors’ and researchers’ work for guiding in research. The systematic literature review does not directly take the content of research authors’ primary work rather, it refracts the information through their lens. The present study thoroughly acknowledges every author’s contribution who has been consulted by using the citation and referencing.
 Lodder, AR., and Zeleznikow, J., 2005. Developing an online dispute resolution environment: Dialogue tools and negotiation support systems in a three-step model. Harv Negot L Rev 10:287.
 Thiessen, E., and Zeleznikow, J., 2004. Technical aspects of online dispute resolution challenges and opportunities. In: Proceedings of the third annual forum on online dispute resolution, Melbourne, Australia, pp 5–6.
 Kumar, R., 2018.Research methodology: A step-by-step guide for beginners. Sage.
Allen, M. (Ed.). (2017).The SAGE encyclopedia of communication research methods. Sage Publications.
 Zhu, M., Sari, A., & Lee, M. M. (2018). A systematic review of research methods and topics of the empirical MOOC literature (2014–2016).The Internet and Higher Education,37, 31-39.
 Snyder, H., 2019. Literature review as a research methodology: An overview and guidelines.Journal of business research,104, pp.333-339.
 Parker, C., Scott, S. and Geddes, A., 2019. Snowball sampling.SAGE research methods foundations.
 Hadi, M.A. and Closs, S.J., 2016. Ensuring rigour and trustworthiness of qualitative research in clinical pharmacy.International journal of clinical pharmacy,38(3), pp.641-646.
This chapter briefly elaborates the data extraction by review of documents from previous literature. The PRISMA flow diagram demonstrates the flow of information from different phases and channels of the systematic literature review. It helped the present research to identify the number of review documents, research articles inclusion and exclusion, and the reason behind exclusion. Also, Critical appraisal skill program (CASP) tools are explained that helped the present research to develop skills through analyzing the research evidence and applying them in practice. Systematic literature is examined using tabulation representation in which key features such as the author’s name, research topics, research methodology employed, and the results of the study are highlighted. Also, key findings of the present research are highlighted at the end.
PRISMA Flow Diagram
The inclusion and exclusion criteria are the main part of the systematic literature review. Inclusion criteria have everything that a study needs to include in a review. Contrary to this, exclusion criteria detail the components which can make the study ineligible hence they need to be excluded from the review. The PRISMA and AMSTAR (assessment of multiple systematic reviews) tools have wide use in exclusion and inclusion checklist criteria (Kelly et al. 2016) . The study explains that AMSTAR tool is applied to evaluate the methodological quality of systematic literature review. Whereas, PRISMA tool’s focus on systematic reviews come through analyzing randomized trial, particularly in the analysis of interventions. Moreover, the PRISMA tool provides evidence-based components used for systematic reviews and meta-analysis. The present research used the PRISMA tool for systematic review because of its fundamental importance to; exhibit quality of reviews, allow researchers to assess the weaknesses and strengths, and allow replicating review technique. Below is the diagram that demonstrates the PRISMA tool used in the present research:
 Kelly, S.E., Moher, D. and Clifford, T.J., 2016. Quality of conduct and reporting in rapid reviews: an exploration of compliance with PRISMA and AMSTAR guidelines.Systematic reviews,5(1), pp.1-19.
Figure 1.2: PRISMA Tool and Paper Selection
After a deep insight and searching through different engines, the present research found research articles matching the keywords “AI use in ODR and its application in the justice system” out of 39 selected research papers and books, the present study found 9 papers (Tyler, 1987; Gómez-Pérez et al. 2006; Zlatanska and Betancourt, 2013; Kaufmann-Kohler and Schultz, 2004; Sharmila, 2020; Tuleutayevna, 2016; Ebner, 2021; Simkova and Smutny, 2019; Van Arsdale, 2015) as replicated cases of discussing the AI use in ODR with a different perspective. Thus, these papers are excluded from the consideration for appearing duplicated. Furthermore, 5 papers i.e. Dourish and Bell, 2011; O’callaghan, 2003; Makridakis, 2017; Moravec, 1998; Bales, 2017 have been excluded because they are found irrelevant to AI application in ODR.
In the second stage further 10 research papers have been excluded because the present research found rare information in these papers relating to the objectives of the study. Therefore, for the final systematic literature review, the present research relied on 4 research books authored by Wilson-Evered and Zeleznikow, 2021; Mustajoki, 2007; Gélinas et al. 2015; Carneiro et al. 2014. These books have been considered because they provide very useful information relating to practical applications of AI in ODR and its use in the justice system. The present study has also considered a research report authored by Keplesky (2007) which also provides useful information relating to the present study objectives. The report provides an insight into the practical implications of AI using examples. For final data synthesis and data extraction the present study considered the following research articles:
|Zeng et al. (2020)||AI; science communication||Contested Chinese dreams of AI? Public discourse about artificial intelligence on WeChat and people’s daily.||To examine the challenges to China’s AI program at the national level.||The study is based on limited data analysis due to heavily censored information on the topic.|
|Lodder and Zelznikow (2005)||AI and Law; AI and dispute resolution; game theory||Developing an online dispute resolution environment: Dialogue tools and negotiation support systems in a three-step model.||To merge the game theory with negotiation techniques to develop a model for the ODR environment.||ODR successful resolution depends on dispute domain i.e. family dispute. Limited implications.|
|Sela (2018)||ODR; arbitration mediation; ODR in Justice||How Automated and Human-Powered Online Dispute Resolution Affect Procedural Justice in Mediation and Arbitration.||To investigate the disputants’ dependency on procedural justice through ODR processes.||Does not explain the issues and challenges associated with ODR tools in the justice system.|
|Ebner and Zeleznikow (2016)||ODR; negotiation; mediation: technology||No Sheriff in Town: Governance for Online Dispute Resolution.||To explore ODR applications in the governance system.||Descriptive study and based on more Narrative analysis.|
|Thompson (2015)||AI; ODR; Justice; access legal system; HCI; pathway||Creating New Pathways to Justice Using Simple Artificial Intelligence and Online Dispute Resolution.||To analyze the design of AI-based systems for non-experts help in the justice domain; AI applications in the justice field||Does not analyze the challenges in the justice domain and ways for AI to counter those challenges.|
|Condlin (2016)||ODR; AI-based software programs;||Online Dispute Resolution: Stinky, Repugnant, or Drab.||To evaluate the challenges in the widespread implementation of ODR.||The research article missed the systematic overview.|
|Surden (2018)||AI; Human and hybrid system; AI and law||Artificial intelligence and law: An overview.||To analyze AI use in the law domain.||It does not cover the issues relating to AI use in law.|
|Meany and Clark (2010)||Computer technology and human interaction; communication; emotions||Humour Theory and Conversational Agents: An Application in the Development of Computer-based Agents.||To evaluate the emotions involved in human and technology interaction; analyze human behavior in adopting technological communication.||The research article missed the systematic overview.|
|Mania (2015)||ODR; mediation; arbitration; e-commerce||Online dispute resolution: The future of justice||To measure the strengths and weaknesses of ODR; to analyze regulatory challenges for ODR.||It analysis the ODR related challenges in general and not for specific fields.|
|Qutieshat (2017)||ODR; communication network; family disputes||Online Dispute Resolution||To evaluate the issues in adopting ODR.||The research article missed the systematic overview.|
Table 1.2: Selected Article Description
Critical Appraisal Skills Program (CASP) Tool
Singh (2013)  defines CASP as a generic tool for appraising the limitation and strengths of any qualitative research. It provides useful assistance to the researchers to extract meaningful, relevant, and valid information from the existing literature, which matches the study objectives. For the present research, the study relied on keywords and appraisal tools that match the present study’s objectives and theme. For further extracting and elaboration, different steps have been followed which led to the completion of the present study task. Following is the figure which shows the steps taken to obtain the relevant research material:
 Singh, J., 2013. Critical appraisal skills programme.Journal of pharmacology and Pharmacotherapeutics,4(1), p.76.
Figure 1.3: CASP Tool
The eligibility criteria for inclusion and exclusion have been discussed in research methodology and PRISMA tool procedure. For publications distribution, the present research considered only the research articles published after 2003 except referring to some articles of 1977 and onwards when analyzing the historical perspective on the topic. For research journals and authors, the present study relied on three main sources Google Scholar, Science Directory, and Emerald Insight. The paper citation has been considered as the main focus and the study considered only those papers that have more than 10 citations.
Synthesis and Data Analysis
Data synthesis is a statistical measure to combine the results of different studies and literature to obtain a qualitative estimate of the overall effect of a specific variable or intervention on the defined outcomes (Park et al. 2018) . It helps the researchers to combine the arguments, ideas, findings, recommendations, and critical reviews of different researchers into a systematic manner. Since the current study is undertaking a systematic literature review, where most of the literature has been found on qualitative work based on arguments and facts. Therefore, rather to combine a particular intervention or variable, the present research focuses on the arguments, ideas, judgments, and critical reviews of the previous researches. Data synthesis has been divided into two sections based on the research objectives i.e. the role of AI application in the resolution of online disputes, its usefulness for the parties, and the challenges involved in the process. The other section analysis the AI applications for ODR in the law domain. Following is a brief discussion of the mentioned objectives as well as it also analyzes the levels of human risks relating to AI use in ODR and provides examples of cases of AI use in ODR:
Role of AI in ODR and Challenges
There has been an increased focus on developing AI-based designs for ODR during the last six years (Zeleznikow, 2021) . The study believes that this focus got more importance due to the COVID-19 spread lately. The prevalence of the pandemic is resulted in more dependency on using ODR and it requires advanced AI systems to address the users’ needs. The study explores a wide range of AI applications in ODR such as communication support, providing counsel, decision making, management, administration, and facilitating the agreements. Lodder and Zeleznikow (2012)  evaluate the use of AI for key ODR techniques such as negotiation, mediation, and arbitration. AI-based program of interactive computer has been developed to provide facilitation to those parties who are in process of negotiation to settle the conflicting matters.
The system can be used by the involved parties or a professional mediator can facilitate the process. The study argues that AI applied in ODR provides a range of investigating techniques to identify and clear the interests of parties involved. Furthermore, it clarifies the tradeoffs, allows party’s satisfaction, and identifies as well as provides optimal solutions. AI-based programs take the information provided by each party and after processing the information it provides existing feasible alternatives to the parties which should be chosen above the involved parties’ proposals. If the program cannot find a feasible alternative, it supports parties to generate counter-proposals. The study by Carneiro et al. (2014)  measures the AI contribution in ODR is solving complicated issues by using the tools of decision support system, expert system, knowledge-based system, and intelligent interfaces, etc.
However, the study argues that the majority of existing ODR applications depend on obtaining information by using the traditional methods. The way this information is stored provides little or no support to AI systems and barriers for the wider implementation of these systems. Therefore, the use of AI-based systems is little due to the inability of handling the models for complex information. Thompson (2015)  finds that AI is a computer-based machine that is unable to understand the involved parties’ feelings and their social requirements. The process of human interaction with computers gives the basis that how humans can utilize and interact with technology. Thus people seeking dispute settlement by using the internet often come across emotional distress.
AI in Law Domain
The study by Lodder and Zeleznikow (2012) explores the AI facilitation for negotiation support in ODR and such advice is very useful in conflict resolution in the law domain. The advice provided by AI emphasis upon justice and does not sheer focus on the disputants’ interests. Thompson (2015) analysis that AI can perform a range of functions and can also guide individuals as per their particular circumstances in the law domain. The study stresses the importance of adopting JPES approach which offers a considerable contribution to dispute settlement in the law and justice field. However, Carneiro et al. (2014) argue that law is not so simple to interpret. The interpretation of rules often creates doubts among legal disputants that lead to clashing and varying interpretations and, varying outcomes.
Thus the study finds that AI comes across the challenge of rules interpretation while working under such a complex legal system. Moreover, AI also faces the challenge to deal with the changing nature of rules, especially in the civil law area. The study determines that when a new case is solved in court, there come more cases of different nature, thus AI cannot adapt to deal with consistently rapidly changing data and information. ODR needs an AI system in the law domain that can be adaptable to the changes in rules and norms. The developers are making efforts to design such systems and making them flexible for storing updated information without generating ambiguities.
Human Risks and Examples of AI-Based Cases
The study by Quest et al. (2018)  identifies the potential risks of using AI in ODR such as biased conclusions drawn by the system based on gender, race, and age. Companies using AI for dispute settlement can also face backlash from clients who have a fear of data manipulation and exploitation especially if the record surveillance is also accessed by the government. The study also provides an example of banks using AI; these banks need to test and analyze their money laundering and fraud monitoring for AI-driven programs to avoid irrational penalizing to any specific group. Similarly, if racial biases inadvertently develop into a system, it can issue false alerts by recognizing individuals to be suspicious or criminal by mistake. Moreover, the workforce of the firm using AI can also become too much dependent on AI tools to catch fraudulent for them. This approach developed in the workforce can build a feeling of comfort in them, therefore, they are expected to miss the regular surveillance and fail to identify the obvious cases. Chandra (2016)  determines that for being affordable more parties can access AI-based programs for dispute settlement which can increase disputes. The system can also have an inability to resolve disputes where cases are different and new for the system.
Quest et al. (2018) find that AI tools help companies to reduce their workforce and automate routine of evaluation helps in cutting the false alerts. The study gives the example of the Royal Bank of Scotland which was saved from $9 million losses due to using an AI-based system for checking transactions relating to small businesses and recognized fake invoices. The study by Trajtenberg (2018)  provides an example of AI use in the law system where expert systems or KBS not merely decide cases but they give explanation and justification of their decisions under a particular rule of law that enables them to reach the outcome. For example under administrative law, AI-based system not only decides that unemployment benefit can be given but it also gives access to the employee for checking purpose. In Ireland, this kind of AI-based system is used by Prosecuting authority to measure the due punishment for imprisonment cases where the penalty is not more than four years.
 Park, N., Mohammadi, M., Gorde, K., Jajodia, S., Park, H. and Kim, Y., 2018. Data synthesis based on generative adversarial networks.arXiv preprint arXiv:1806.03384.
 Zeleznikow, J., 2021. Using Artificial Intelligence to provide Intelligent Dispute Resolution Support.Group Decision and Negotiation, pp.1-24.
 Lodder, A.R. and Zeleznikow, J., 2012. Artificial intelligence and online dispute resolution.Online Dispute Resolution: Theory and Practice A Treatise on Technology and Dispute Resolution, pp.73-94.
 Carneiro, D., Novais, P. and Neves, J., 2014. Conflict resolution and its context.Cham: Springer.
 Thompson, D., 2015. Creating new pathways to justice using simple artificial intelligence and online dispute resolution.IJODR,2, p.4.
 Quest, L., Charrie, A., du Croo de Jongh, L. and Roy, S., 2018. The risks and benefits of using AI to detect crime.Harv. Bus. Rev. Digit. Artic, pp.2-5.
 Chandra, G.R., 2016. Cyber Space for Universal Peace: The Contribution of Online Dispute Resolution.IUP Law Review,6(4).
 Trajtenberg, M., 2018.AI as the next GPT: a Political-Economy Perspective(No. w24245). National Bureau of Economic Research.
Conclusion and Recommendations
This chapter presents the conclusion and policy recommendations on the role of AI and its application in ODR. The chapter also presents the limitation of the present study.
There are advantages as well as disadvantages of using AI in ODR. Internet culture is inclined to anonymity; hence ODR preserves anonymity and settles the dispute besides. However, when disputes are settled through using AI-based systems, the involved parties feel a lack of human interaction and emotions. Customers also fear manipulation and exploitation of their information from online platforms. In the law domain, the absence of any authority figure makes cheating easy for the involved parties. However, there is a range of advantages of using ODR such as it provides access to dispute resolution in a fast and affordable manner especially in the law domain. Due to the use of AI techniques for the solution of problems, the decisions are made by AI programs that are more rational and objective due to less human emotions involvement. AI contribute for ODR to be more effective to deal with more complex issues such as multi-issue, multi-agreements, and multi-groups.
However, the present study analyzes that the communication and interaction tools used in current ODR processes are very old-fashioned. In dispute settlement cases the involved parties mostly provide data and information using web forms. ODR systems work without providing an advanced form of interaction where users can be provided support and assistance throughout the process. Thus the study believes that employing AI techniques can substantially develop complete automated systems for ODR. The study also observes that ODR mediation and negotiation methods provide a range of implementation; however, arbitration adoption is limited.
The potential scope of ODR is not limited keeping in view the people more dependency on communication through electronic means and power of electronic interaction. Awareness of dispute settlement methods makes people more independent in managing their conflicts by using ODR means. The use of AI in ODR makes the dispute settlement procedures more standardized and day-to-day tasks can be dealt effectively through software programs without relying on humans. Therefore, the software program scope for use in the future is high in dispute resolution.
Issus can be complicated to resolve and difficulty is faced in resolving the dispute especially in the justice domain where rules need interpretation. On such occasions, the developers should wisely design the program to assign a part of the process to the human performer and another part to the software performer. Software is not programmed with simplicity to make rational decisions for disputants from time to time to compromise or find a moderate settlement of a complex legal conflict. A software performs what it is commanded, thus emotional consideration, thinking, moral evaluation, legal analysis, and political assessments are the functions that are not performed by the software in dispute resolution. These factors make ODR come under criticism. For the solution of legal issues, AI-based systems come across certain challenges such as interpretation of rules, dealing with changes in rules, and handling a new case that is very different from previously dealt cases. Thus AI-based designs need to be developed further for adaptability to store information and keeping the information updated.
The study concludes that ODR techniques should be based on clear provisions for the disputants which classify the legal relationship based on particular characteristics and kinds of claims. Electronic tools are necessary for creating a legal network. Such a platform facilitates the automatic settlement of issues regardless of consuming time on court hearings. The concept of JPES demonstrates the significant assistance for non-experts to use AI-based advanced systems in the justice domain to obtain the resolution of disputes. Keeping in view the users’ particular circumstances, these expert systems offer expert knowledge to perform a range of functions such as decision making, guidance, and providing advice. JPES techniques are not much technically sophisticated as compared to existing AI systems in the law field. However, it is an advanced tool that takes part in the area of ODR widely.
There are many challenges to be countered in AI development and researchers will consistently develop the effective tools and techniques that will enable the AI-based systems to perform more efficiently and become accessible to people. However, the study recommends that developers should not develop highly complex and advanced systems rather the focus should be on developing such systems which can be used by non-expert individuals who have no or minimum knowledge about the domain particularly such initiatives are required in the law domain. Such tools will provide quick and efficient settlement of disputes and will provide fast and effective reach out to the justice system. Such expert systems should be hybrid by merging the rule-based and case-based systems for simplicity of the design. It can lead to a more effective solution to the disputes in a transparent manner. The information can be accessed easily through this system which can efficiently enhance the dispute settlement procedures and can bring more satisfaction to the parties involved.
Limitations of the Study
Besides its positive and deep insight contribution to the literature, the current study has some limitations. The current study briefly investigated the highlighted research objectives through the systematic literature review, but as mentioned that the research articles have been collected through snowball sampling. Therefore, continuing the same research objectives with more efforts on searching relevant literature on the relevant topic and relevant case study area will give a more accurate and true picture of the topic. Secondly, the number of papers used in systematic literature in the current study was limited, therefore the extended number of literature may give more insight to the research area.