Smart Courts and Digital Justice: A Governance Framework for Responsible Artificial Intelligence in Judicial Decision-Making

Abstract

Artificial intelligence (AI) is now used in many parts of digital court systems, including e-filing, registry scrutiny, transcription, translation, legal research, document review, and case-management support. This review examines how AI can help smart courts while protecting the values that make adjudication legitimate: legality, fairness, reason-giving, accountability, judicial independence, and public trust. The study uses a qualitative scoping-review design. It draws on peer-reviewed legal scholarship, empirical studies, technical literature, institutional frameworks, regulatory materials, and official court-technology documents published mainly between 2018 and 2026. The review shows that AI is most appropriately deployed in administrative, informational, and assistive tasks that improve speed, access, language inclusion, and document management. Uses that predict outcomes, influence bail or sentencing, recommend risk scores, or draft judicial reasons create higher risks, including bias, opacity, privacy intrusion, automation bias, and diffusion of responsibility. The article proposes a human-centred and rights-based governance framework in which AI operates as accountable court infrastructure rather than an autonomous adjudicator. Smart courts therefore require not only technology, but also clear policies, independent validation, audit trails, contestability, data-protection safeguards, vendor controls, and judicial capacity building. AI can strengthen digital justice only when it supports, and does not replace, human judicial reasoning.