Artificial Intelligence in the Indian Judiciary: Balancing Technological Efficiency with Constitutional Values

ABSTRACT

AI has gradually changed the way courts and legal systems across the world operate,[i] helping create court systems that are more effective than ever by enhancing case management and legal research as well providing increased access to justice[ii]. Currently, the Indian judiciary is adopting the use of AI through various projects including Project SUPACE[iii], Project SUVAS[iv] and the E-Courts Mission Mode Project.[v] Although using AI in the judiciary has potential benefits[vi] with respect to better administration of justice and reduced backlog of cases, the introduction of AI into the judicial system[vii] presents serious constitutionally and ethically related issues, including the independence of the judiciary, fairness, transparency, accountability, and possible algorithmic bias.[viii] The purpose of this paper is to examine critically how AI has been used in the Indian judicial system and whether AI based adjudication is consistent with the constitutionally guaranteed rights of the Indian Constitution[ix]. It will provide a comparative look at international developments pertaining to the use of AI[x] in the judiciary in other countries and will discuss the need for judicial decision-making to include a human voice. This paper also contains a comparative analysis of AI developments in international jurisdictions. The analysis also emphasizes the importance of incorporating a human element into judicial decision-making. The authors argue that AI can be used as an effective tool[xi] for judgement but that relying only on algorithmic decisions will weaken the fundamental principles of constitutional justice.[xii] The authors that AI may supplement the capability of judges[xiii]; however, it must be implemented consistent with constitutional principles of morality, transparency, accountability, and discretion.

Keywords:

Artificial Intelligence; The Judiciary; Judicial Independence; Algorithmic Bias; Constitutional Law; Article 14; Article 21; AI Ethics; SUPACE; Judicial Technology.

 

Introduction

The Indian Judiciary has many long-standing problems. The judiciary has a lot of excessive numbers of pending court cases and not enough judges available to take up those cases.[xiv] There are also many backlog cases due to way of processing – just too slow. Also, the lack of adequate judicial infrastructure is overwhelming. Data has shown[xv] that millions of cases remain pending in multiple courts and that the failure of judicial systems to provide timely adjudication of these pending matters violates the Constitutional right of every citizen to have access to effective, expeditious justice. In light of this environment, the use of technological advancements is starting to enhance the efficiency of justice delivery systems significantly. A specific area of technology that will have a tremendous effect on the future of almost every legal system globally will be Artificial Intelligence (AI)[xvi]. AI is already having an influence on legal systems around the globe. Artificial intelligence can be defined as a computer system that performs tasks normally requiring human cognitive ability including reasoning, data analysis, pattern recognition, language processing and predicting what the future will hold. The use of AI tools [xvii]and software in the legal field has been on the rise and these tools are used extensively for legal research, contract review, document analysis, transcript creation, case management, and predicting judicial outcomes. Countries such as the U.S., China, and Estonia[xviii] already have various types of AI being used for their court systems. India has also begun taking some preliminary steps to incorporate AI into its court system[xix]. Three examples from the Government of India are: e-Courts Mission Mode Project[xx] (to create an electronic court record system),SUPACE (Supreme Court Portal for Assistance in Court Efficiency)

SUVAS (Supreme Court Vidhik Anuvaad Software), and others. All of these examples demonstrate that the Indian judiciary is being receptive and willing to embrace technology into its court system.The development and rapid integration of digital technology into court systems is being greatly influenced by the use of AI. With the recent upsurge in virtual hearings as a result of the ongoing pandemic[xxi], court systems have been able to implement and use various types of digital technologies much more rapidly than they would have otherwise been able to[xxii]. This has made it clear that there is an urgent and pressing need for modernization of judicial administration. In many ways, the pandemic has exposed deficiencies in the current judicial administration systems.

The incorporation of AI into the decision-making process[xxiii] of the judiciary poses serious legal and ethical issues,[xxiv] particularly with respect to issues related to judicial independence, transparency, accountability, the presence of bias in algorithms, and the right to a fair trial.[xxv] All of these will be major points of contention in any discussion about the use of AI in the courts. Although AI will enhance judicial efficiency[xxvi] and decrease the burden of work upon judges[xxvii], the administration of justice involves more than just analysis of data; it requires a level of compassion, the ability to make balanced decisions, knowledge of the Constitution, ability to exercise discretion, and understanding of social context, and it is unclear that if any of these characteristics could ever be fully replicated by machines[xxviii]. Therefore, the discussion about the use of AI in the judiciary is more than a mere technological discussion; it is also very much a constitutional issue. One of the key questions regarding the use of AI in the judiciary[xxix] is whether AI could act to assist judges without compromising the fundamental principles of fairness, accountability, and independence which are the basis for the Indian judiciary[xxx].

The meaning of AI in a Judicial System

AI in a Judicial Systems means the use of advanced computer algorithms and machine learning technology to assist courts and legal professionals with the completion of their judicial and administrative function. AI systems can analyze large amounts of data regarding laws, identify patterns within the data they analyzed, predict outcomes based upon the data they are analyzing, and provide analytical assistance to the user in an incredibly short time period, using machine learning.

AI is utilized in various ways in legal systems globally. For instance, lawyers can use predictive analytics to evaluate their chances of winning a court case. Lawyers may also find it easier to find legal literature with automated legal research tools than they would using traditional methods. Some US states, as an illustration, have implemented risk-assessment algorithms to aid inting should an individual be released on bail and what the punishment should be. For example, in the US, the COMPAS algorithm is utilized to assess the likelihood of a defendant repeating a crime after being arrested. Additionally, “intelligent courts” have also been established in China, which utilize AI to manage and resolve cases through an online interface. Lastly, small claims courts in Estonia have investigated the potential for utilizing AI to assist in processing cases where the amount of money being sought is minimal.

Generally, judicial systems can classify the use of AI into two broad categories. The first category is applications that are assistive to the courts and include tasks such as research, transcription, translation, scheduling, and case management. The second category is applications that adjudicate cases and include any AI application that would directly influence the result of a judicial proceeding. The application of assistive technologies is widely viewed as positive, while the application of adjudication technologies is highly debated, with concerns about fairness and constitutionality of using AI in adjudication processes. Yet, gradual growth of artificial intelligence’s role in courts requires consideration of potential legal and constitutional consequences for those technologies.

Integration of AI in the Indian Judiciary

Over the last ten years, advancements in technology have greatly impacted Indian courts. The E-Courts Mission Mode Project was initiated under the National e-Governance Plan to digitise court records, make judicial record available to the public, and improve the efficiency of courts. Currently, e-filing, online case status updates, virtual hearings and digital records are all common practice within the Indian courts.

One of the first AI initiatives is SUPACE, an AI software that provides judges with assistance in conducting legal research and analysing their cases. Judges are expected to use SUPACE as an aide to their own judgements — this system helps judges accurately locate and organise relevant facts, identify applicable precedents, and efficiently collate the legal information needed to make a ruling on a case. The Supreme Court of India has also launched another AI initiative called SUVAS, which is a software program that leverages machine learning to produce translations of legal documents into regional languages. This initiative aims to help litigants better understand the proceedings of courts and the ruling that were issued, in reference to the rights in regional languages. In addition, the advent of COVID-19 has hastened the adoption of information and communication technologies (ICT) within the courts. During lockdowns, virtual hearings became a necessary option, thus providing proof that ICT can assist courts more effectively with judicial administration; there are many examples of this type of engagement. As of now, the Indian judiciary views the use of artificial intelligence (AI) as an assistive technology for court/legal processes, but does not approve of using AI to fully automate the judiciary’s adjudication process (and the resulting determination). The Indian judiciary has begun to experiment with some form of AI, such as voice transcription tools that use AI technology and/or digital case management systems powered by AI. Despite some positive advances, no legal authority has provided to AI the ability to independently decide cases. The judiciary has repeatedly stated that AI should assist judges; not replace them. Judicial reasoning could be impacted by the ever-increasing use of AI recommendations; thus, there is a need for scrutiny of this new reliance upon AI to render its recommendations, relative to judicial conduct, under the U.S. Constitution.

Constitutional and Legal Issues- The Right to Fair Trial and Due Process

The right to a fair trial is seen as being encapsulated within Article 21 of the Indian Constitution.

The Supreme Court in Maneka Gandhi v. Union of India has held that any process that affects an individuals right to personal liberty, must be conducted through a fair, just and reasonable process. AI-assisted judicial systems create questions regarding procedural fairness. This is largely due to the fact that the algorithms used to power an AI use highly complex processes that the average person will not be able to understand, nor defend against those algorithms used to make judicative decisions shall create a level of difficulty and frustration for individuals wishing to challenge the validity of AI-generated recommendations resulting in a particular judgment. Automated systems could violate the principle of natural justice if parties do not have access to review how algorithmic decisions were made. The lack of transparency could also be a violation of constitutional guarantees to fair processes and due process.

Judicial Independence

Judicial independence is one the essential foundations of a constitutional democracy, as judges are expected to decide cases independently, free from any external influence or pressure. The increased reliance upon AI-generated recommendations may, therefore, indirectly infringe upon or undermine the judicial discretion and autonomy.

Although AI is being developed with the intent to support judges, it is possible that the judiciary will gradually become accustomed to relying upon algorithmic evaluations of evidence, particularly in high-volume civil case litigation. If this becomes the case, then judicial reasoning could be diminished and the judicial process may be more susceptible to biases produced by AI technologies through the influence of the output of AI on the judicial judgment and the ultimate consequence for a party litigating before a judge. Judges are required to engage in the analysis of social realities, human conduct, moral conduct, and constitutional values in making their decisions in adjudication. These factors cannot wholly be expressed using statistical data or predictive models. Therefore, if technology begins to dominate judicial reasoning, then judicial independence will be compromised.

Algorithmic Bias and Discrimination

One of the most significant issues with AI systems is the presence of bias in algorithmic decision-making. AI systems operate by developing statistical models based on historic data; If this data reflects social biases and/or discrimination that has already existed and occurred throughout the course of history, then AI’s output will reproduce the same statistical patterns of bias and/or discrimination as in the original data that the AI drew its training examples, based upon. Biased data sets can lead to unequal impacts on marginalized communities based on their religious, gender, caste and/or socio-economic status; thus predictive algorithms in criminal justice systems could wrongly classify some communities or individuals based on historical crime statistics rather than their present day actions.

All of this discrimination is unconstitutional and violates Article 14 of the Constitution which guarantees everyone equal treatment by the law as well as equal protection under the law. The Supreme Court of India held that an arbitrary classification would violate the legal principles of equality in the case of State of West Bengal and Anwar Ali Sarkar. The threat that AI could be used to institutionalise bias therefore poses an enormous threat to the rule of law and equal justice.

 

Privacy and Data Protection Issues

AI systems necessarily require access to large amounts of data to train and operate correctly. AI systems in the justice system use great amounts of sensitive personal information when they process things like criminal history, financial accounts or personal communications related with people being court.

In Justice K.S. Puttaswamy v. Union of India, the Supreme Court ruled that the right to privacy is a fundamental right protected by Article no. 21 of The Constitution. There could be violations against individual privacy rights through excessive data collection or through algorithm monitoring in the judicial system as such processes are typically unobstructed by safeguards. The Supreme Court in Anuradha Bhasin v. Union of India made clear that protecting constitutional rights for individuals must continue to be maintained when individuals are using digital media or using technology. Also highlighted in this case were proportionality and procedural protections while regulating digital technologies. The Supreme Court also affirmed the need to protect individual constitutional rights against ambiguous, and vague and ambiguous, excessive technological regulation, as demonstrated in Shreya Singhal v. Union of India. Given these principles, when evaluating AI-based judicial systems, it is critical to use these principles in connection with evaluating AI-based judicial systems.

Therefore, when implementing AI in the courts, there should be strong laws protecting personal data as well as overall enforceable regulatory standards relating to appropriate use of that data. The Benefits of Artificial Intelligence (AI) in Judicial Administrations. Despite concerns over Constitutionality. However, there are real benefits from using AI with Judicial Administrations; such as: One, AI can reduce the legal research workload of Judges by providing faster search capabilities to locate relevant case law, statutes, and legal principles that will speed up judicial time and improve efficiency.

Two, AI based case Management Systems will assist the judicial system by establishing a better method of organizing, scheduling, and prioritizing matters that require immediate attention, as well as providing methods to streamline the administrative functions of the Judiciary System. In addition, providing People with tools, such as Automated Transcription and Translation, will improve these individuals’ ability to access the courts, regardless of their language(s). Three, AI will assist in promoting more consistent applications of Justice and Judgment by helping to identify cases that have contradicted each other, as well as providing assistance in establishing consistent principles for the application of laws. AI will be able to help to reduce human errors in the repetitive tasks that occur in the Administrative aspects of the judicial system. The use of AI systems in the Judicial System will be an effective tool in bringing efficiency to the Institutions of the Judicial Systems, as well as improving the Public’s ability to access the Judicial System.

Judicial Reasoning Being Replaced By Machines – The Risks Associated With Replacing Human Judges/Decision Makers With Machines. Judicial Administrations will save a court system by assisting in the operation of the court systems. However, if Human Judges/Technicians are replaced by Devices, the concept of Justice and how it is served in a particular society will be changed and/or denigrated. Judicial Decision makers make decisions based on how an individual will interpret that action, and/or, how to define as an honourable, ethical, and/or fair manner in society. Justice is not a function of mathematical calculation of information patterns. Judicial courts are to uphold constitutional morality, protect the vulnerable, and provide fair results in all situations. This includes scenarios in which strict compliance to legal structure cause unjust circumstances. Judges can use empathy, discretion, and moral judgement; machines cannot.

Public confidence in a court’s decision-making ability are dependent on the court’s ability to resolve disputes through the thinking of unbiased human minds that are capable of understanding the circumstances of each case. More unreasonable automation of judicial procedure may create the image of ‘mechanical’ or ‘artificial’ decisions that lack compassion or human empathy. Thus, AI should not be viewed as a replacement for judicial decision-making and should be used only as a support tool for judicial decision-making.

 

 

 

Recommendations and Reforms for the Indian Judiciary

The use of AI in the Indian Judiciary should be done in a manner that is both responsible and prudent:

1. There should be no room for AI in any decision-making process (ie, as an adjudicator); therefore, AI should be limited only to assistance or administratively based functions. The last judicial authority should remain with the human judge.
2. Every AI system in use by the court must be transparent to the user so they have the right to know how their AI-assisted recommendations were developed, what factors the algorithm considered, and be able to challenge the recommendation if it impacts a judicial proceeding.
3. There should be independent bodies set up to regulate how and when AI is used by the judicial court system, along with regular audits conducted to eliminate algorithmic bias within the system.
4. There should be strict protections that would provide for the personal privacy and confidentiality of all litigants for all data gathered, utilized and stored by AI systems within the court system. All judicial officers must be provided with appropriate technology training in order to utilize AI tools intelligently, however the deployment of artificial intelligence in the judicial system must conform to the core values of fairness, equality, accountability, and independence throughout the stages of AI being implemented.

Artificial Intelligence is capable of substantially changing the administration of justice to improve efficiencies, broaden access to justice, and enhance the management of cases. India, with an overburdened judicial system and an enormous backlog of cases, could take advantage of technology to strengthen the delivery of justice within its borders. SUPACE and SUVAS are two initiatives supporting India’s move toward a modernized system of justice through responsible use of technology.  In contrast, although one could quantify justice with numbers or mathematically compute justice through numbers using an algorithm, there is a significant component of empathy in judicial decision-making; the application of constitutional moralities in keeping with appropriate ethical principles and the complexity of being human are all things that machines will find difficult. An overreliance on AI jeopardizes the independence of the judiciary, procedural justice/fairness, equality before the law and public faith in the judicial system.

Consequently, AI should serve an adjunct function and not replace judges when rendering decisions on justice. Implementing AI in the Indian judicial process should comply with constitutional standards, transparency, accountability, and have built-in oversight from humans who are exercising reasonable discretion. Although technology will be meaningful in assisting with the administration of justice, the response from the bench to administer justice still lies with the humans (judges) who must measure the will of the law against compassion/emotional needs of humanity. 

REFERENCES

1. Constitution of India, Article 14.
2. Constitution of India, Article 21.
3. Constitution of India, Preamble.
4. Maneka Gandhi v. Union of India, (1978) 1 SCC 248.
5. Justice K.S. Puttaswamy v. Union of India, (2017) 10 SCC 1.
6. State of West Bengal v. Anwar Ali Sarkar, AIR 1952 SC 75.
7. Anuradha Bhasin v. Union of India, (2020) 3 SCC 637.
8. Shreya Singhal v. Union of India, (2015) 5 SCC 1.
9. State v. Loomis, 881 N.W.2d 749 (Wis. 2016).
10. Richard Susskind, Online Courts and the Future of Justice (Oxford University Press 2019).
11. Richard Susskind, Tomorrow’s Lawyers: An Introduction to Your Future (2nd ed., 2017).
12. Cary Coglianese & David Lehr, “Regulating by Robot: Administrative Decision Making in the Machine-Learning Era,” 105 Georgetown Law Journal 1147 (2017).
13. Frank Pasquale, The Black Box Society: The Secret Algorithms That Control Money and Information (Harvard University Press 2015).
14. Danielle Keats Citron & Frank Pasquale, “The Scored Society: Due Process for Automated Predictions,” 89 Washington Law Review 1 (2014).
15. Brian Z. Tamanaha, Beyond the Formalist-Realist Divide: The Role of Politics in Judging (Princeton University Press 2010).
16. Mireille Hildebrandt, “Law as Computation in the Era of Artificial Legal Intelligence,” in Speaking Law to Power (2020).
17. Roscoe Pound, The Spirit of the Common Law (1921).
18. Lon L. Fuller, The Morality of Law (Revised ed. 1969).
19. H.L.A. Hart, The Concept of Law (3rd ed. 2012).
20. Shoshana Zuboff, The Age of Surveillance Capitalism (2019).
21. OECD, Artificial Intelligence in Society (2019).
22. UNESCO, Recommendation on the Ethics of Artificial Intelligence (2021).
23. European Commission, Ethics Guidelines for Trustworthy AI (2019).
24. John McCarthy, “What is Artificial Intelligence?” Stanford University (2007).
25. NITI Aayog, National Strategy for Artificial Intelligence #AIForAll (2018).
26. Indian Judiciary Annual Report 2023–24, Department of Justice, Government of India.
27. Ministry of Electronics and Information Technology, Government of India, Digital Personal Data Protection Act, 2023.
28. Supreme Court of India, SUPACE (Supreme Court Portal for Assistance in Court Efficiency).
29. Supreme Court of India, SUVAS (Supreme Court Vidhik Anuvaad Software).
30. Ministry of Law and Justice, Government of India, E-Courts Mission Mode Project.
31. Supreme Court of India, Guidelines and Vision Document for Phase III of the E-Courts Project.
32. United Nations Secretary-General, Roadmap for Digital Cooperation (2020).

WEBOGRAPHY

1. Supreme Court Observer
2. Supreme Court of India
3. E-Courts Mission Mode Project
4. European Commission Digital Strategy
5. Stanford University AI Papers
6. NITI Aayog Official Website
7. UNESCO Digital Library
8. United Nations Official Website
9. Ministry of Electronics and Information Technology India
10. Department of Justice India

 

[i] Richard Susskind, Online Courts and the Future of Justice.

[ii] OECD, Artificial Intelligence in Society

[iii] Supreme Court of India, SUPACE (Supreme Court Portal for Assistance in Court Efficiency), Supreme Court Observer, https://www.scobserver.in.

[iv] Supreme Court of India, SUVAS (Supreme Court Vidhik Anuvaad Software), https://main.sci.gov.in

[v] Indian Judiciary Annual Report 2023–24, Department of Justice, Government of India.

[vi][vi] Danielle Citron & Frank Pasquale, The Scored Society

[vii] Frank Pasquale, The Black Box Society

[viii] Cary Coglianese & David Lehr, Regulating by Robot

[ix] INDIA CONST. art. 21.

[x] European Commission, Ethics Guidelines for Trustworthy AI

[xi] Brian Tamanaha, Beyond the Formalist-Realist Divide

[xii] Roscoe Pound, The Spirit of the Common Law

[xiii] Lon L. Fuller, The Morality of Law

[xiv] Indian Judiciary Annual Report 2023–24

[xv] Indian Judiciary Annual Report 2023–24

[xvi] OECD, Artificial Intelligence in Society 67–82 (2019).

[xvii] Richard Susskind, Online Courts and the Future of Justice

[xviii] Richard Susskind, Tomorrow’s Lawyers: An Introduction to Your Future 45–58 (2d ed. 2017).

[xix] NITI Aayog, National Strategy for Artificial Intelligence #AIForAll (2018), https://www.niti.gov.in

[xx] Richard Susskind, Online Courts and the Future of Justice 102–10 (Oxford Univ. Press 2019).

[xxi] Mireille Hildebrandt, Law as Computation in the Era of Artificial Legal Intelligence, in Speaking Law to Power 3, 7–12 (2020).

[xxii] Supreme Court of India, Guidelines and Vision Document for Phase III of the E-Courts Project

[xxiii] Cary Coglianese & David Lehr, Regulating by Robot

[xxiv] Frank Pasquale, The Black Box Society

[xxv] Danielle Citron & Frank Pasquale, The Scored Society

[xxvi] Brian Tamanaha, Beyond the Formalist-Realist Divide

[xxvii] Roscoe Pound, The Spirit of the Common Law

[xxviii] Lon L. Fuller, The Morality of Law

[xxix] European Commission, Ethics Guidelines for Trustworthy AI

[xxx] Brian Tamanaha, Beyond the Formalist-Realist Divide