‍INTRODUCTION

Alternative Dispute Resolution (ADR) constitutes a collective framework of mechanisms, including arbitration, mediation and conciliation, through which disputes are resolved outside of the conventional judicial process. Arbitration means the submission of disputes by mutual agreement of the parties to one or more arbitrators, whose decision is binding upon the parties who submitted to arbitration[1]. By opting for arbitration and other ADR processes, parties choose a private, party-driven system of dispute resolution rather than traditional court litigation, where there is minimal to no party autonomy. Disputes are referred to arbitration in accordance with an arbitration clause incorporated in the original contract, or via a separate arbitration agreement or  through judicial referral.

The ADR ecosystem worldwide is currently undergoing a transformation with the incorporation of artificial intelligence (hereinafter referred to as “AI”). The application of AI within ADR has the potential to reshape contemporary dispute resolution practice. A major contribution of AI lies in increasing the efficiency and expediting the proceedings by the automation of routine and repetitive tasks, streamlined case management and data-driven support in decision making. Such technology reduces procedural delays and contributes to reducing the burden on an already overburdened Indian legal system, thus resulting in cost savings and improved access to ADR mechanisms.

Hence, the question which arises for consideration is whether the use of AI in the ADR processes, particularly in reference to the use of predictive analytics, is feasible in contemporary times.

This project analyses the role of AI in ADR, particularly in reference to predictive analytics for predicting probable case outcomes, its application in online dispute resolution and the use of assistive technologies to support ADR processes and critically evaluates these possibilities in light of the existing Indian legal framework governing ADR.

ANALYSIS

AI in ADR: The Indian Perspective

India has undertaken progressive measures to encourage the adoption of technology, including AI, within the broader framework of ADR. The judiciary has played a key role in legitimising the use of technological tools in arbitration and allied ADR mechanisms. The Supreme Court, in Shakti Bhog Foods Ltd. v. Kola Shipping Ltd.[2] and Trimex International FZE Ltd. v. Vedanta Aluminium Ltd.[3], has upheld the validity of using electronic modes of communication in an arbitral proceeding and has also recognised arbitration agreements concluded through emails and other similar electronic exchanges.

In Grid Corporation of Orissa Ltd. v. AES Corporation[4], the Supreme Court further clarified that notices regarding appointment of arbitrators may be communicated through e-mail and are not necessarily to be served in physical or written form upon the concerned party. The Court further allowed the enforcement of arbitration agreements formed through electronic correspondence even in the absence of physical signatures. This judicial approach later played a key role in expansion of Section 7(4)(b)[5] of the Arbitration and Conciliation Act, 1996[6], to expressly recognise “electronic means” as a legitimate mode for the formation of arbitration agreements.

Additionally, institutional and procedural reforms have strengthened the integration of technology in ADR. Under the Mission Mode Project[7], the introduction of e-filing systems has allowed the digital submission of documents/pleadings and the electronic signing of documents[8]. Such initiatives show the judicial recognition of technology as an enabler of efficiency, accessibility, and procedural economy in dispute resolution.

These developments show the judicial and legislative intent to streamline ADR in India by technological integration. The facilitation of electronic communication and digital documentation has rendered the ADR process more cost-effective, expeditious, and user-friendly. Moreover, advanced applications of AI, including predictive analytics and data-driven decision-support systems, hold the potential to strengthen ADR mechanisms by enhancing case management and enabling informed outcome assessment.

How Artificial Intelligence Influences ADR

Predictive Analytics

One of the most significant domains within Alternative Dispute Resolution where AI may be effectively deployed is predictive analytics for the assessment of probable case outcomes. By analysing outcomes of similar prior disputes, AI-driven predictive systems enable parties and their legal representatives to undertake informed risk assessment and to structure their pleadings and arguments accordingly. Such analytical assistance supports more rational decision-making within ADR processes and promotes early resolution wherever applicable.[9]

However, the benefit of predictive analytics depends upon the availability of a comprehensive and reliable data repository consisting of prior arbitral awards and ADR outcomes. If an arbitral institution or designated body maintains structured and anonymised databases of past decisions, AI systems can identify patterns relating to legal issues, procedural choices, composition of tribunals, and factual matrices. Although no predictive model has yet been developed specifically for arbitration or ADR, some algorithms have demonstrated considerable accuracy in predicting outcomes. Particularly, academic studies conducted by institutions such as the University College London, the University of Pennsylvania, and the University of Sheffield have successfully predicted outcomes of cases concerning Articles 3, 6, and 8 of the Convention for the Protection of Human Rights and Fundamental Freedoms[10] with an accuracy of approximately 79 per cent[11]. These studies demonstrate the capacity of AI-based systems in analysing judicial reasoning and forecasting outcomes with a high degree of reliability.[12]

Furthermore, predictive analytics platforms, such as the Ravel Law, compiles extensive datasets by examining the work product of various law firms, thus, generating more structured and contextually relevant insights. The judicial acceptance of such technology is can also be evidenced by referring to the decision in Pyrrho Investments Ltd. v. MWB Property Ltd.[13], wherein the court permitted use of predictive coding software for certain document’s disclosure. In the aforesaid case, the claimant was required to review more than three million documents and predictive coding, as distinct from traditional linear review, enabled the prioritisation of documents through algorithmic analysis, thereby enhancing efficiency and reducing the burden of manual examination of the said documents.[14] This decision recognised that AI assisted tools improve decision making efficiency and thus, mitigate the impracticality of human review in cases involving voluminous data.

The effectiveness of the predictive analytics within ADR is, however, inherently dependent upon the quality, volume, and accessibility of relevant data. ADR mechanisms operate within a confidential and privatised framework and consequently, several leading arbitral institutions, including the International Chamber of Commerce and the Hong Kong International Arbitration Centre, either do not publish the arbitral awards or limit their publication to anonymised summaries.[15] While such confidentiality safeguards party autonomy and privacy, it simultaneously constrains the development of robust predictive models due to absence of comprehensive datasets.

In order to address this limitation, emerging technologies have adopted certain alternative approaches. Tools such as the GAR Arbitrator Research Tool compiles anonymised data obtained from parties and practitioners across multiple jurisdictions, covering ADR processes.[16] This method preserves confidentiality while facilitating data-driven analysis and predictive assessment.[17] Such innovations demonstrate that it is possible to balance out the foundational principles of confidentiality in any ADR process with the growing demand for analytical transparency and efficiency.

From an Indian perspective, use of predictive analytics in ADR remains feasible; however, it would necessitate forming a structured, anonymised and a legally compliant database of past awards and outcomes. However, any such initiative must be aligned with data protection standards, privacy norms and the consensual nature of ADR. If implemented within these constraints, AI-based predictive systems can considerably enhance the effectiveness, predictability and credibility of arbitration and other ADR mechanisms in India.

Online Dispute Resolution

Online Dispute Resolution (ODR) is a noteworthy technological development within the broader landscape of ADR, which uses information and communication technology to enable dispute resolution through digital platforms, thus eliminating or substantially reducing the need for the physical presence of the parties. The integration of artificial intelligence within ODR platforms includes decision support systems, negotiation support tools, and automated advisory or counselling mechanisms and all of which facilitates structured online negotiation and informed decision making.[18]

AI-enabled ODR mechanisms offer procedural and economic advantages, such as by minimising costs associated with travel, venue infrastructure and logistical arrangements. ODR enhances the affordability and accessibility of ADR processes. This is particularly useful in cases where the dispute involves parties from multiple jurisdictions and where in-person hearings (and logistics of timing and location) may be impossible. The ODR platforms leverage algorithms to facilitate administrative processes, help produce standard documents, monitor case timelines and other procedural flows, making the whole process more efficient.

Some ODR systems also offer features such as computer-assisted counselling that gives parties coaching and guidance based on their entered information, enabling them to gain a sense of the nature and potential  paths of action with respect to their dispute. Internationally, a number of countries have started to formally accept and institutionalise ODR as a legal method for resolving disputes. In particular, African nations are in the process of acceding to international conventions and  developing domestic legislation supporting the use of online dispute resolution (ODR) within their justice administration systems. Similarly, the ASEAN Committee on Consumer Protection has issued Guidelines with respect to the Online Dispute Resolution and some of their member states, such as Indonesia and Thailand, have already operationalised national ODR frameworks.

India also has a growing institutional and judicial recognition of ODR. Courts have acknowledged the use of ODR not only at preliminary stages of a proceeding but also in substantive phases of dispute resolution. The Supreme Court of India has also highlighted the relevance of technology-driven justice delivery and has also constituted an e-Committee to prioritise the development and implementation of digital dispute resolution mechanisms.

Despite progressive policy initiatives, the Indian legal system continues to exhibit reliance on physical and formalistic procedures. This is evident from decisions such as Coastal Marine Constructions & Engineering Ltd. v. Garware Wall Ropes Ltd.[19], wherein the Supreme Court has held that an arbitration agreement is unenforceable unless the requisite stamp duty has been duly paid. Although the introduction of e-stamping and online payment mechanisms by the Central Government has simplified compliance, however, certain State level procedural requirements, such as physical annexation of e-stamp certificates, continue to obstruct the seamless integration of end-to-end online dispute resolution.

Assistive Technologies

A broad spectrum of generalised assistive technologies enabled by artificial intelligence plays a key role in the procedural functioning of ADR framework. Tools based on natural language processing (NLP) and machine learning algorithms help parties, tribunals and institutions in managing disputes more efficiently. NLP integrates statistical modelling, machine learning, deep learning, and computational linguistics to analyse, interpret, and generate human language in both written and spoken form.

In international commercial arbitration and cross-border ADR, document translation poses a significant procedural challenge as parties, at many times, operate in different linguistic environments; arbitration agreements may be drafted in one language, while the seat of arbitration or the jurisdiction of enforcement, as the case may be, may require another. In the Indian context, the Hon’ble Supreme Court has created the Supreme Court Vidhik Anuvaad Software (SuVAS), which is an artificial intelligence-based translation tool that translates judgments, orders and legal documents from English to various Indian languages.[20] These efforts promote process efficiency and access in ADR, so are a further step toward lean dispute resolution.

This use of new technologies is also true for international languages and further illustrates the global nature of ADR. The India International Arbitration Centre has also signed an MOU with Bhasha Interface for India (BHASHINI) for use of Artificial Intelligence-powered language translation and interpretation software during arbitral proceedings. This MoU provides an opportunity for linguistic assistance and a level-playing field to all parties, irrespective of the States they belong and in any language spoken by such party.[21] This is an illustration of how AI-driven assistive technology increases inclusivity and procedural fairness and leads to gaining insight into disputes, which can then be resolved quickly, hence also supporting predictive analytics.

CONCLUSION

Artificial intelligence can change the ADR landscape in the present day. The use of predictive analytics, in particular, serves a dual function within ADR as it helps parties and legal practitioners evaluate litigation and settlement risks, formulate strategies, and assess the probable trajectory of a dispute even before it formally enters arbitration or other ADR mechanisms. Moreover, the utility of such predictive tools is currently limited to the amount and quality of data, which inevitably tends to be scarce as a result of (i) confidentiality and privacy constraints associated with arbitration and other ADR processes; and (ii) because parties only tend to disclose such minimum information that they are compelled to provide.

Another major progress is the integration of AI into Online Dispute Resolution. At present, ODR is not limited to the field of online arbitration and is perceived as a solution capable of containing disputes, avoiding disputes and resolving disputes by using integrated digital solutions. Furthermore, ODR also enables the processing of differences before they become a formal dispute at a later stage.

Moreover, AI-driven assistive technologies further aid in the procedural efficiency of ADR. Document review, transcription and translation services and case management tools save time, expense and administrative hassles and also provide broader accessibility and better inclusivity in the ADR process. These technologies leave intact the non-workplace canon of arbitration and other ADR mechanisms while enhancing their working functioning without altering them at their basic level, i.e. consensual and adjudicatory. Although AI can greatly improve the efficiency, consistency and depth of analysis in ADR procedures, its role continues to be an adjunct rather than a replacement for human judgment. For this reason, the role of artificial intelligence in ADR must be understood as one that is complementary to procedural efficacy and informed decision making, but at the same time is technologically limited, legally regulated and still requiring human oversight.

REFERENCES

[1] World Intell. Prop. Org., What Is Intellectual Property? (2009).

[2] (2009) 2 S.C.C. 134 (India).

[3] (2010) 3 S.C.C. 1 (India).

[4] 2005 SCC OnLine Ori 78 (India).

[5] Arbitration and Conciliation Act, No. 26 of 1996, § 7(4)(b), India Code (1996).

[6] Arbitration and Conciliation Act, No. 26 of 1996, India Code (1996).

[7] Ministry of Elecs. & Info. Tech., Gov’t of India, Mission Mode Projects.

[8] eCommittee, Supreme Court of India, eCourts Project – Phase II: Objective Accomplishment Report as per Policy Action Plan Document.

[9] Gülüm Bayraktaroğlu-Özçelik & Až. Barış Özçelik, Use of AI-Based Technologies in International Commercial Arbitration, 12 Eur. J.L. & Tech. 1 (2021).

[10] Convention for the Protection of Human Rights and Fundamental Freedoms arts. 3, 6 & 8, Nov. 4, 1950, 213 U.N.T.S. 221.

[11] Nikolaos Aletras, Dimitrios Tsarapatsanis, Daniel Preoţiuc-Pietro & Vasileios Lampos, Predicting Judicial Decisions of the European Court of Human Rights: A Natural Language Processing Perspective, PeerJ Comput. Sci. (2016).

[12] Said Gulyamov & Mokhinur Bakhramova, Digitalisation of International Arbitration and Dispute Resolution by Artificial Intelligence, 9 World Bull. Mgmt. & L. 79 (2022).

[13] [2016] EWHC 256 (Ch).

[14] Oliver Browne & Hayley Pizzey, Pyrrho Invs. Ltd v. MWB Prop. Ltd: A Landmark Decision on Predictive Coding in E-Discovery.

[15] Kathleen Paisley & Edna Sussman, Artificial Intelligence Challenges and Opportunities for International Arbitration, 11 N.Y. Disp. Resol. Law. 1 (2018).

[16] Nikolaos Aletras, Dimitrios Tsarapatsanis, Daniel Preoţiuc-Pietro & Vasileios Lampos, Predicting Judicial Decisions of the European Court of Human Rights: A Natural Language Processing Perspective, PeerJ Comput. Sci. 93 (2016).

[17] Austrian Yearbook on International Arbitration 2019 (Wolters Kluwer 2019).

[18] Ass’n of Southeast Asian Nations, ASEAN Launches Guideline on Online Dispute Resolution for Consumers.

[19] (2019) 9 S.C.C. 209 (India).

[20] Supreme Court of India, Press Release (Nov. 25, 2019), https://main.sci.gov.in/pdf/Press/press%20release%20for%20law%20day%20celebratoin.pdf.

[21] Mahnoor Waqar, The Use of AI in Arbitral Proceedings, 37 Ohio St. J. on Disp. Resol. 3 (2022).