The Supreme Court has released draft regulations on the use of artificial intelligence (AI) in courts, allowing lawyers and litigants to use AI tools in legal work while requiring disclosure of such use. The proposed framework makes clear that the power to decide questions of law, fact and justice will remain exclusively with judges.

The draft Regulations for Use of Artificial Intelligence in Courts, 2026, released by the Supreme Court's AI Committee for public consultation until June 20, would apply to the Supreme Court, High Courts, subordinate courts, tribunals and statutory adjudicatory bodies nationwide.

Under the proposed framework, AI may be deployed for legal research, citation verification, drafting assistance, translation, transcription, case management, scheduling, record management and judicial administration. The regulations also encourage the use of AI-powered tools to improve accessibility and assist litigants in navigating court services.

At the same time, the draft draws clear boundaries around the technology. AI systems would be barred from deciding cases, determining bail, passing sentences, evaluating witness credibility, predicting future conduct or influencing judicial deliberations. Courts would also be prohibited from relying on opaque "black-box" systems in matters affecting rights or personal liberty, including AI-based risk-scoring tools used to assess the likelihood of reoffending or absconding.

A key feature of the proposal is a mandatory disclosure regime. Any party or legal representative using AI in preparing pleadings, documents, submissions or evidence would be required to inform the court at the time of filing. Courts may also seek details about the AI system used, the extent of assistance provided and the verification measures adopted to ensure accuracy.

The regulations place responsibility on litigants and lawyers for the content they submit. If AI-generated material is found to be false, fabricated, misleading or inaccurate, the person filing it would remain fully accountable and could not cite the AI-generated nature of the material as a defence.

The framework is anchored in the principle of "human primacy", stipulating that AI must function only as an assistive tool and cannot replace the independent exercise of judicial authority. Final responsibility for decisions on facts, law and justice would continue to rest with judges and court officials, who likewise cannot rely on AI errors or hallucinations to justify incorrect decisions.

Recognising the risks posed by AI hallucinations, the draft requires verification of AI-generated outputs before their use in judicial processes.

To oversee implementation, the proposal envisages a permanent apex AI body at the Supreme Court level, supported by AI Committees and dedicated secretariats across courts. It also calls for annual audits of AI systems, maintenance of AI registers and incident databases, cybersecurity safeguards, compliance with data protection laws, and regular training for judges, lawyers and court staff.

Notably, the draft adopts a policy favouring responsible AI adoption, stating that courts should actively explore technologies capable of improving access to justice, reducing delays and enhancing efficiency.