Source: Digital Statecraft

The Supreme Court of India’s draft "Regulations for the Use of Artificial Intelligence in Courts, 2026" is presented as a milestone that recognises growing AI use in judicial administration and aims to ensure technology strengthens the administration of justice. The draft is built on the intuition that AI may assist but humans should remain responsible for judicial decisions; the author questions whether regulating who makes decisions is the right organising principle and argues instead for regulating the legitimacy of decisions themselves. Every consequential public decision follows a protocol—facts established, rules identified and interpreted, discretion exercised, reasons recorded—and AI changes who performs parts of that protocol, not the protocol itself.

The author proposes a mandatory, standard "Decision Receipt" for every consequential public decision that records material facts, their sources, rules and precedents applied (with versions), the reasoning linking facts to rules, points of discretion, participating systems and people, the final accountable authority, and mechanisms for review and appeal. The receipt should be both machine-readable for automated audit and readable in plain language for affected citizens. This instrument would improve explainability, auditability and contestability, help appellate courts reconstruct reasoning, enable continuous AI auditing, reveal inconsistent treatment across cases, and address the "prior-access paradox" and procedural fairness by recording how information entered the decision process and what assistance was provided. The author recommends that the Supreme Court’s regulations shift from insisting on human subordination of AI to requiring standard Decision Receipts for every consequential public decision.