Addressing the 'Black Box' Problem: Ensuring Transparency and Explainability in Ai Systems for Indian Criminal Justice

Main Article Content

Kritika Goyal, Shreya

Abstract

Artificial Intelligence (AI) is being applied within criminal justice systems across the world, including in India, to improve efficiencies in policing, evidence analysis and adjudication. However, the implementation of these AI-driven tools has leads to an urgent need to consider transparency and accountability. This is particularly important in regard to the "black box" problem - that is, the inability to make clear to society the logic that is active within the algorithm. In the context of India, this lack of transparency is a problematic feature of the criminal justice system compounded by systemic bias, infrastructure, and due process protections within the constitution. If these systems lack transparency, it is reasonable to be concerned that AI systems could similarly replicate existing society biases, lead to unjust consequences and diminish public faith in the legal system more broadly. In this paper, we engage in a critical discussion of the implications of the black box problem in AI application to the Indian criminal justice system. The paper investigates how a lack of explainability can hinder the rights of accused persons, the standards of evidence in criminal proceedings, and judicial reasoning. The paper also looks at comparative perspectives from the United States and the European Union, which take a legal-ethical approach to the expectation of explainability and accountability of algorithms. The paper advocates for India to create strong regulatory interventions, including an algorithmic audit, disclosure controls, and recommendation of Explainable AI (XAI) as part of the legal processes in India. By establishing transparency and explainability as principles, India can enjoy the progress and positive aspects of technological innovation, and also continue to shelter the constitutional promise of justice. Ultimately, we need to ensure AI systems are designed to be interpretable and accountable to preserve the goals of democratic values and the legitimacy of the Indian court system in the administration of justice.

Article Details

How to Cite
Kritika Goyal, Shreya. (2026). Addressing the ’Black Box’ Problem: Ensuring Transparency and Explainability in Ai Systems for Indian Criminal Justice. European Economic Letters (EEL), 16(1), 1210–1218. Retrieved from https://eelet.org.uk/index.php/journal/article/view/4261
Section
Articles