Suggestions on a Preliminary Draft of ‘Regulations for Use of Artificial Intelligence (AI) in Courts, 2026’ submitted on June 5, 2026
- Communications Team
- Jun 5
- 2 min read
On June 5, 2026, a comprehensive, formal submission containing structural suggestions and precise textual amendments to the preliminary draft of the Regulations for Use of Artificial Intelligence (AI) in Courts, 2026 was officially filed before the Secretariat of the Artificial Intelligence Committee of the Supreme Court of India.
The extensive feedback document was developed jointly by Abhivardhan and Ayush Chandra, Co-founder of LegalTechPolicy.com, in response to the public notice issued by the Hon’ble Supreme Court on June 4, 2026, inviting stakeholder consultations.
Core Interventions and Structural Reforms Proposed
The memorandum outlines critical statutory and philosophical vulnerabilities within the proposed regulatory framework, offering precise parliamentary drafting redrafts to ensure the rules preserve constitutional protections while modernizing judicial administration. The key interventions include:
The Quasi-Judicial Taxonomy: Critique of Regulation 3(1)(b) and 19 for lumping highly consequential acts like "defect scrutiny" and "issuance of summons" into a purely catch-all administrative definition. The submission demonstrates that automated systems operating without human-in-the-loop verification in these areas risk violating procedural fairness, and proposes a distinct "Quasi-Judicial" statutory classification.
Bifurcation of Sensitive Judicial Data (SJD): A comprehensive framework to split SJD into Tier-1 (Adjudicatory SJD) and Tier-2 (Administrative SJD) to eliminate the operational paralysis of a monolithic classification. The draft proposes absolute on-premise air-gapping for evidentiary core data, while programmatically aligning registry metadata with the Digital Personal Data Protection Act, 2023.
Reversing the "Innovation Bias" Loophole: Structural redrafting of Regulation 17(2) to establish a strict, non-derogable value hierarchy ensuring that the principles of Human Primacy, Rule of Law, and Proportionality at all times supersede any statutory presumption of rapid AI adoption.
Banning Synthetic Evidence: Re-versioning Regulation 20(1)(h) to alter the baseline from mere transparent disclosure to a strict statutory prohibition against introducing AI-generated synthetic data or algorithmic reconstructions as independent substantive evidence.
Proactive Right to Human Intervention: Introduction of an upfront, non-prejudicial procedural right for litigants to opt out of automated translation or transcription tools in favor of traditional human-operated frameworks.
Mitigating Legacy Transitional Immunity: Tightening Regulation 41 to introduce a tiered, risk-based prioritization timeline and a strict 90-day non-compliance sunset clause to prevent unsafe legacy court software from running indefinitely during review phases.
Document Access
To facilitate academic discussion, public transparent engagement, and legal analysis among practitioners, technology vendors, and civil society groups, the complete feedback text submitted to the Apex Court has been made public.
The complete memorandum, featuring the precise statutory redrafts and a model demonstration of the systemic vulnerabilities of programmatic anonymization against metadata re-identification, is available for immediate download.
Comments