The Supreme Court's draft AI-in-courts rules: what the 2026 regulations mean for your practice
On 3 June 2026, the Supreme Court of India's Artificial Intelligence Committee released the Draft Regulations for the Use of Artificial Intelligence in Courts, 2026 — the first time the apex court has attempted to set rules for how AI may operate inside the Indian judicial system. For advocates wondering whether AI in law is hype or substance, this is a clear signal: it is here, and it is being governed.
What the draft actually permits
Rather than banning or rubber-stamping AI, the draft takes a measured, use-case approach. Permissible uses include:
- Case management within the registry and court administration.
- Automated transcription of proceedings — but with mandatory human review and an accuracy certification.
- Translation of judgments and legal documents, subject to human verification.
- Legal research assistance.
- Accessibility tools for persons with disabilities.
The recurring theme is augmentation under supervision — AI assists, humans remain accountable.
The principles worth internalising
Two ideas in the draft matter for every practitioner, well beyond the courtroom:
- Transparency and explainability (Regulation 7). AI systems used in court are expected to meet high standards of transparency — you should be able to understand how an output was produced.
- Restrictions on high-risk uses. The draft limits AI in applications that affect personal liberty. Decisions about a person's freedom are not to be outsourced to a model.
These are not just rules for the registry. They are a template for how the profession should think about adopting AI at all: as a tool whose outputs you can explain and must verify — never as an oracle.
What this means for how you use AI
If you are using AI for drafting, research, or summarising, the court's own framework gives you a sound professional standard to hold yourself to:
- Verify everything. The draft insists on human review of AI transcription and translation. Apply the same discipline to AI-assisted research and drafting — check the citations, read the authorities, own the output.
- Keep judgment human. AI can surface a precedent or draft a first cut. The legal judgment about whether it fits your matter is, and should remain, yours.
- Mind confidentiality. Feeding client facts into the wrong tool can be a data-protection problem (see our piece on the DPDP Rules). Use AI tools that respect where your data goes.
Why "bring your own key" matters here
This is exactly why Lawisense takes a deliberately cautious approach to AI: it stays optional, it is never the last word, and where AI features are used, they are designed to keep you in control of your own keys and your own data — not to route confidential client matters through systems you cannot see into. The Supreme Court's draft validates that posture. The future of AI in Indian law is not autonomous machines deciding cases; it is well-governed tools making careful lawyers faster.
The regulations are still in draft and will evolve. But the direction is set: AI in Indian courts will be transparent, supervised, and human-accountable. That is a future advocates can work with.
Sources: The Leaflet — Explained: SC Draft AI Regulations 2026, Mondaq — India's Supreme Court framework to govern AI in courts.