AI IN THE JUDICIAL SYSTEM: POSSIBLE USES AND ETHICAL CONSIDERATIONS
The widespread adoption and use of Artificial Intelligence (AI) in different sectors has opened the world up to its infinite possibilities of use. AI has a domino effect on technological advancements – increasing system and industry efficiency and transforming the ecosystem of different sectors. This introductory piece explores:
- how AI is perceived and used in the judicial system,
- the challenges and ethical considerations that must be evaluated in its adoption and use, and
- whether, within the African context, judicial systems are ready and open to the use of AI.
Artificial Intelligence in the Judicial System.
Technology has played a vital role in creating a foundational basis for the adoption of AI. The digitalization of Court systems in different jurisdictions can be cited as the most widespread use of technology. “The process of digitising the courts began at the end of the third industrial revolution. Now, the revolution of information technology provides an excellent opportunity to transform the judicial system into a mesmerizingly quick, efficient, and high-quality array of services available to all citizens and residents.”1 The digitization of court systems creates a foundational basis upon which available data can be used to identify the spaces in the judicial system where AI application can have significant impact.
Introducing AI to the justice system promises to improve procedural and administrative efficiency, aid in decision making processes for judges, lawyers and litigants, and further predict outcomes consistent with past precedents.2 AI is in use in many judicial systems globally.3 The EU, US, and UK are applying AI interventions in their respective judicial systems. In Estonia for example, the Estonian Ministry of Justice has designed a ‘robot judge’ to adjudicate small claims’ disputes of less than €7,000 (about $8,000). The first pilot test was initiated to resolve contract disputes; the Ministry of Justice hopes to eventually expand the use of the ‘robot judge’ to other claims.4
In Brazil, an AI tool called VICTOR is used to conduct preliminary case analysis to reduce the burden on the court. The tool supports the Brazilian Supreme Court by providing analysis of cases using document analysis and Natural Language Processing (NLP). 5 Singapore’s Court system has adopted speech translation systems. The system utilises neural networks trained with language models and domain specific terms to transcribe court hearings in real time allowing the review of oral testimonials instantaneously by judges and parties.6 In Austria, AI is used for document management such as anonymization of court documents and as a digitization assistant for analogue files.7 The public prosecutions office of Buenos Aires (Argentina) and the Constitutional Court of Colombia utilize a tool called Prometa to predict the outcome of cases; the tool has so far recorded a 96% success rate. It is also able to identify urgent cases within large volumes of files in under 2 minutes where it would normally take a human being an average of 96 days8
The US and the UK have utilised AI to design risk assessment tools. These tools are quite controversial, as they raise a number of ethical challenges, specifically on issues around bias, transparency, 9 The Strategic Subject List (S.S.L) introduced in Chicago was intended to predict individuals are likely to be involved in gun violence.10 Another tool, COMPAS – Correctional Offender Management Profiling for Alternative Sanctions has been used to assess recidivism risk11 and inform parole sentencing decisions. A similar tool is also being used in the UK to predict which criminals are most likely to be a repeat offender, the information is used to suggest the kind of supervision the defendant ought to receive in prison.12 The controversial nature of these tools arises from the inherent bias in the algorithms and training datasets used to build the AI platform.
The use cases reflect a diverse application of AI in the judicial system. We must, however, also consider the ethical challenges that the use of AI presents, especially in the case of automated decision making. Transparency and explainability are important where AI driven technologies are adopted. Lack of transparency in regards to algorithms used, often as a result of legal protection of trade secrets, is likely to undermine the rule of law. “The principle of the rule of law requires that rules are publicly declared with prospective application and possess the characteristics of generality, equality and certainty. An important procedural dimension of the rule of law, which is of particular importance is the effective capability to contest decisions. Algorithms are data driven and in this context the lack of transparency would undermine the rule of law as a procedural safeguard to discern, foresee, understand and contest decisions.”13 A lack of understanding of the algorithms by the users as they are not trained to understand the technicalities of the systems, creates a situation that is likely to be detrimental to the judicial system. Majority of the users are not trained to understand the algorithms and their operations, this leads to a situation where the makers of the AI solutions have more information than the users of the AI solutions. Identifying possible biases that would affect the outcome of cases becomes difficult, and the lack of knowledge as to the inputs and outputs that facilitate the functioning of the system undermines the judicial norm of reasoned orders.14 The use of AI in a public institution, like the judiciary, must be mindful of the pre-existing, inextricable social contexts and dynamics that affect them.15
The EU, through the European Commission for the Efficiency of Justice (CEPEJ), adopted the European Ethical Charter on the use of Artificial Intelligence in Judicial Systems and their Environment.16 The Charter was developed for public and private stakeholders responsible for the design and deployment of AI tools and services that involve the processing of judicial decisions and data. The charter introduced five principles to be considered in the use of AI in judicial systems and their environment. It identifies the uses of AI in the judicial systems of EU member states and presents an overview of the importance of open data policies relating to judicial decisions in judicial systems of EU member states. The Charter also touches on AI and legal reasoning, how AI can be applied in civil commercial and administrative justice; the protection of personal data, and the potential limitations of predictive justice tools. In addition, the charter also specifies uses of AI in European judicial systems that can be encouraged, uses requiring considerable methodological precautions, uses to be considered following additional scientific studies, and uses to be considered with the most extreme reservations.
The five principles identified in the charter are the following:
Respect for fundamental rights: this principle requires ensuring that the design and implementation of artificial intelligence tools and services are compatible with fundamental rights.
Non-discrimination: Specifically prevent the development or intensification of any discrimination between individuals or groups of individuals
Quality and security: With regard to the processing of judicial decisions and data, use certified sources and intangible data with models conceived in a multi-disciplinary manner, in a secure technological environment.
Transparency, impartiality and fairness: Make data processing methods accessible and understandable, and authorize external audits.
“Under user control”: Preclude a prescriptive approach and ensure that users are informed actors and in control of their choices.
Although the Charter was developed for EU member state countries, the degree of development of access to justice in AI instruments differs from state to state – starting from random distribution of cases, processing of judicial decisions and data by AI, tools of predictive justice up to the solving of cases by a robot. The principles serve as a reference point on the fundamental values and methodological requirements for the creation and development of algorithms for other jurisdictions.17
The African Context
AI has been widely adopted in African countries in various sectors: finance, health, education, access to public services, etc. The judicial sector on the continent is also taking a keen interest in the use of AI. Access to justice in Africa remains fundamentally important in the protection, recognition and enforcement of human rights.18 Digitalization of court systems in Africa is one of the first steps that has been taken in the judicial systems embracing technology. This has improved efficiency in the judicial system, specifically in addressing the challenges of accessibility, long delays, accountability, transparency, and corruption.19
Although the judicial systems in the African jurisdictions are still at the foundational stages in the adoption of technology, and the use of AI is not as prevalent as it is in the EU, USA and the UK, law firms in Africa have, nevertheless, embraced AI tools to improve their legal service delivery. AI platforms such as LawPavillionPrime, launched in 2016, gives in-depth analysis of the strengths and weaknesses of legal positions and authorities by generating statistical analysis, historical data, precedential value ratings, conflicting judgments, locus classicus, statutory or literary authorities, and opinions. TIMI, launched in 2018, is Nigeria’s first artificial intelligence legal assistant. It assists lawyers with legal research, litigation, opinion drafting, provides notes with legal authorities, and gives a step-by-step guide on drafting and filing court processes. 20
In trying to identify avenues of AI application in the judicial system, researchers at CIPIT, identified four possible areas of use (in their research paper) of AI in the Kenyan Environmental and Land Court. They noted that AI could be utilized in the specialized court in legal research for land dispute cases, speech recognition and transcription, predictive analysis on case duration and dismissals and online dispute resolution. This demonstrates the possibility of actual utilization of AI in the judicial systems of different jurisdictions in Africa. The differences in how each judicial system functions will however require research and the identification of areas where the use of AI can be most efficient and effective. This will further require involvement of all stakeholders, judicial officers, advocates, and litigants.
Ongoing digital transformations suggest that the adoption of AI technologies is inevitable. AI is here to stay, as such it will be imperative to ensure that while acknowledging all the benefits it presents, consideration is given to the challenges its use presents particularly when used in the judicial system. This remains imperative as the rule of law is the cornerstone of democracies all over the world.
1 Annet Numa, ‘Artificial intelligence as the new reality of e-justice.’ (e- Estonia).<<https://e-estonia.com/artificial-intelligence-as-the-new-reality-of-e-justice/>>
2 Reiling, A.D. (Dory), ‘Courts and Artificial Intelligence.’ International Journal for Court Administration
2020, 11(2), p.8. DOI: <http://doi.org/10.36745/ijca.343>
4 Eric Nillier, ‘Can AI Be a Fair Judge in Court? Estonia Thinks So’ (Wired, 25 March 2019)
5 Daniel Becker and Isabela Ferrari, ‘VICTOR, the Brazilian Supreme Court’s Artificial Intelligence: a beauty or a beast?’ <https://sifocc.org/app/uploads/2020/06/Victor-Beauty-or-the-Beast.pdf>
6 Michelle Chiang, ‘State Courts and A*STAR’s Institute for Infocomm Research (I²R) Collaborate to Develop Real-time Speech Transcription System for Use in Courts’ (State Courts, Singapore, 14 December 2017) <www.a-star.edu.sg/docs/librariesprovider12/press-release-docs/media-release_statecourts-develop-speech-transcription-system-with-astar.pdf>
7 Georg Stawa, ‘Artificial Intelligence – How is Austria approaching AI integration into judicial policies’ (Federal Ministry of Constitutional Affairs, Reforms, Deregulation and Justice Wien, 22 June 2018) <https://rm.coe.int/how-is-austria-approaching-ai-integration-into-judicial-policies-/6808e4d81>
8 Irma Isabel Rivera, ‘The implementation of new technologies under Colombian law and incorporation of artificial intelligence in judicial proceedings’ (International Bar Association, 5 November 2020) <https://www.ibanet.org/article/14AF564F-080C-4CA2-8DDB-7FA909E5C1F4>
9 Ameen Jauhar, Vaidehi Misra, Dr. Arghya Sengupta, Dr. Partha P. Chakrabarti, Dr. Saptarishi Ghosh, Dr. Kripabandhu Ghosh. ‘Responsible Artificial Intelligence for the Indian Justice System.’ (VIDHI, Center for Legal Policy, April 2021) <<https://vidhilegalpolicy.in/wp-content/uploads/2021/04/Responsible-AI-in-the-Indian-Justice-System-A-Strategy-Paper.pdf>>
10Jeff Asher, Rob Arthur, ‘Inside the Algorithm That Tries to Predict Gun Violence in Chicago’s(The New York Times, 13 June 2017) <https://www.nytimes.com/2017/06/13/upshot/what-an-algorithm-reveals-about-life-on-chicagos-high-risk-list.html>
11 Recidivism risk is the likelihood that an individual (either formerly incarcerated and/or under supervision of a justice agency) will commit a crime or violate the conditions of his/her supervision.
12 Ameen Jauhar, Vaidehi Misra, Dr. Arghya Sengupta, Dr. Partha P. Chakrabarti, Dr. Saptarishi Ghosh, Dr. Kripabandhu Ghosh. ‘Responsible Artificial Intelligence for the Indian Justice System.’ (VIDHI, Center for Legal Policy, April 2021) <<https://vidhilegalpolicy.in/wp-content/uploads/2021/04/Responsible-AI-in-the-Indian-Justice-System-A-Strategy-Paper.pdf>>
13 Emre Bayamlıoğlu, Ronald Leenes, ‘The ‘rule of law’ implications of Data-Driven Decision-Making: A Techno-Regulatory Perspective.’ (Law, Innovation and Technology, 2018)
14Ameen Jauhar, Vaidehi Misra, Dr. Arghya Sengupta, Dr. Partha P. Chakrabarti, Dr. Saptarishi Ghosh, Dr. Kripabandhu Ghosh. ‘Responsible Artificial Intelligence for the Indian Justice System.’ (VIDHI, Center for Legal Policy, April 2021) <<https://vidhilegalpolicy.in/research/responsible-ai-for-the-indian-justice-system-a-strategy-paper/>>
15 Ameen Jauhar, Vaidehi Misra, Dr. Arghya Sengupta, Dr. Partha P. Chakrabarti, Dr. Saptarishi Ghosh, Dr. Kripabandhu Ghosh. ‘Responsible Artificial Intelligence for the Indian Justice System.’ (VIDHI, Center for Legal Policy, April 2021)<<https://vidhilegalpolicy.in/research/responsible-ai-for-the-indian-justice-system-a-strategy-paper/>>
16 European Ethical Charter on the use of Artificial Intelligence in Judicial Systems and their Environment.(2018)<<https://rm.coe.int/ethical-charter-en-for-publication-4-december-2018/16808f699c>>
17 Irina Moroianu, Petru Emanuel, ‘Implementation of the European Ethical Charter on the use of Artificial Intelligence in Judicial Systems and their Environment.’ (International Law Review, 2019) <http://internationallawreview.eu/fisiere/pdf/23_Zlatescu__Supliment_Law_Review_SRDE.pdf>>
18 Thabo Magubane, ‘Technology and Law: The Use of Artificial Intelligence and 5G to Access the Courts in Africa.’ (Young African Leaders Journal of Development, Vol 3, Art 11, 01-01-2021) <https://digitalcommons.kennesaw.edu/cgi/viewcontent.cgi?article=1097&context=yaljod>
19 Logan Finucan, Erika Barros Sierra, Namita Rajesh, ‘Smart Courts: Roadmap
for Digital Transformation of Justice in Africa.’ (Access Partnerships, 21 August 2018) <<https://pic.strathmore.edu/wp-content/uploads/2019/03/PIC_WP_Smart_Courts_Roadmap.pdf>>
20 Florence Ogonjo, Joseph Theuri Gitonga, Dr. Angeline Wairegi, ‘Utilizing AI to Improve Efficiency of the Environment and Land Court in the Kenyan Judiciary : Leveraging AI Capabilities in Land Dispute Cases in the Kenyan Environmental and Land Court System.’<http://ceur-ws.org/Vol-2888/paper9.pdf>