Supreme Court Draft AI Regulations 2026: New Rules for Artificial Intelligence in Indian Courts

The Supreme Court of India has published a draft regulatory framework titled the Regulations for Use of Artificial Intelligence (AI) in Courts, 2026, marking the first comprehensive attempt to govern how artificial intelligence may be used across the Indian judicial system. The draft AI regulations, released on June 3, 2026, have been prepared under the supervision of the Supreme Court's AI Committee, and the Court has invited comments from lawyers, technologists, and members of the public until June 20, 2026. For a judiciary that has spent the past year confronting AI-generated fake case citations and experimenting with AI-assisted translation and transcription, the draft is a significant moment. It signals that AI will not be banned from Indian courtrooms, but it will be supervised.

What the Draft Regulations Cover

The draft regulations are not limited to the Supreme Court. They are intended to govern the deployment of AI in the judicial, adjudicatory, and administrative functions of the Supreme Court, the High Courts, tribunals, and statutory bodies exercising adjudicatory powers. This breadth matters. Tribunals such as the NCLT, consumer commissions, and tax tribunals handle enormous caseloads and are precisely the forums where AI-assisted tools are most likely to be deployed for case management and document processing. The framework articulates a presumption in favour of responsible AI adoption, which means the starting point is not prohibition but regulated use, with judicial independence, transparency, accountability, data privacy, and human oversight identified as the governing values.

AI Cannot Judge: The Assistive-Only Principle

The most important principle in the draft is a clear red line: no AI system shall perform the function of adjudication or sentencing. Every AI system deployed in the court ecosystem must function solely in an assistive capacity and must not supplant or compromise the independent exercise of judicial authority. In practical terms, this means a judge may use AI tools for legal research, summarisation of voluminous records, translation of judgments and evidence, or transcription of oral proceedings, but the decision itself, including questions of guilt, liability, bail, and sentence, must remain a human act. This mirrors the approach emerging in several jurisdictions worldwide, where courts have permitted AI as an efficiency tool while treating the judicial function itself as non-delegable.

Mandatory Disclosure of AI Use by Lawyers and Litigants

For practising advocates, the most consequential feature of the draft is a mandatory disclosure requirement. Lawyers and litigants who use AI tools while preparing pleadings, documents, submissions, or evidence will be required to disclose that use. The regulations permit AI to be used for legal research, citation checks, drafting assistance, document summarisation, translation, transcription, and case management functions, but the use must be transparent.

The context for this rule is well known to the profession. Indian courts have recently dealt with instances where AI tools generated citations to judgments that simply do not exist, and the Supreme Court has treated the citation of fictitious cases as a serious matter touching professional misconduct. A disclosure regime puts the responsibility where it belongs: the lawyer who signs the pleading remains accountable for everything in it, whether a human or a machine produced the first draft.

A Permanent Apex Body to Supervise AI in the Judiciary

The draft regulations propose the creation of a permanent Apex Body at the Supreme Court to supervise, regulate, and promote the responsible integration of AI within the judicial system. This body is to be tasked with developing standards, approving AI tools for use in courts, coordinating with the High Courts, monitoring compliance, and publishing annual governance reports. The institutional design suggests a centralised certification model: rather than each court or tribunal independently adopting whatever AI product a vendor offers, tools would pass through an approval process. For legal technology companies building products for the Indian market, this approval pathway will likely become the single most important regulatory gate.

How to Submit Comments on the Draft

The Supreme Court has invited comments and suggestions on the proposed regulations through the Member Secretary of the AI Committee. Stakeholders, legal professionals, technology experts, and members of the public can submit their comments by email to office.regcc@sci.nic.in until June 20, 2026. Given how directly these rules will shape daily litigation practice, bar associations, law firms, legal aid organisations, and legal tech companies all have a strong interest in participating in the consultation. Useful feedback is likely to focus on the practical mechanics: how disclosure should be made in pleadings, what level of AI assistance triggers the obligation, how approved tools will be certified and updated, and how data fed into AI systems will be protected when it includes sensitive case material.

Key Takeaways

The draft Regulations for Use of Artificial Intelligence (AI) in Courts, 2026 establish three pillars: AI may assist but never adjudicate, AI use in court filings must be disclosed, and a permanent Apex Body at the Supreme Court will approve and supervise AI tools across courts and tribunals. The framework adopts a presumption in favour of responsible AI adoption rather than prohibition. Lawyers should begin auditing their own AI workflows now, because once the disclosure requirement takes effect, undisclosed AI use in pleadings could carry professional consequences. The consultation window closes on June 20, 2026, and the final shape of the regulations may change based on the feedback received.