The Supreme Court of India is making strides in accessibility by utilizing AI for translating judicial documents into vernacular languages.

The Delhi High Court's judges-to-population ratio is alarmingly low, with less than one judge per million people, exacerbating the existing backlog of cases.

With 75% of prisoners in India being undertrials, the backlog is projected to take over 300 years to resolve without significant intervention.

To combat these challenges, AI is being introduced to streamline court processes and reduce delays while preserving judicial authority.

Adalat AI, co-founded by Utkarsh Saxena, provides machine learning solutions designed to help courts manage their workloads more effectively.

The Adalat AI application allows judges to dictate orders, which are then automatically transcribed into legal documents, significantly reducing manual paperwork.

This application has been trained on over 100,000 court orders in 11 languages, ensuring it accommodates diverse Indian legal terminology and accents.

AI tools like 'Ask Junior' are also aiding lawyers by summarizing judgments and providing concise legal insights, thereby enhancing their efficiency in court.

As of late April 2025, 'Ask Junior' has successfully summarized over 14,800 judgments with a high accuracy rate, catering to a growing subscriber base.

Moreover, the AI application is integrated into judicial training programs, featuring a feedback system to ensure ongoing improvement.

Currently, Delhi's courts are grappling with a staggering backlog of nearly 1.5 million cases, including 1.29 million criminal and 213,000 civil pending cases.

Currently implemented in 3,000 courtrooms across eight states, Adalat AI aims to expand its presence to 50% of Indian courtrooms by the end of 2025.