India’s Courts Integrate AI Tools to Tackle Backlog Amid Bias Concerns

AI Integration in Indian Courts

India’s judiciary is adopting artificial intelligence (AI) tools to manage its extensive case backlog, which includes over 50 million unresolved cases. Systems such as SUPACE, SUVAS, and Adalat.AI are being integrated into court operations to address delays, with the e-Courts Phase III project allocating ₹7,210 crore for AI and digitization initiatives. SUPACE, developed with IIT Madras, supports judges with legal research and multilingual accessibility, while SUVAS provides AI-driven translation. Adalat.AI, a speech-to-text tool, generates real-time transcripts for Supreme Court proceedings, though its use is restricted to administrative tasks in Kerala’s subordinate courts. These tools aim to automate non-adjudicative functions, easing administrative burdens on an overburdened judiciary.

Global Context and Bias Concerns

“'The moral judgment of the law cannot be delegated to a machine.'”

The integration of AI in Indian courts aligns with global efforts, including those in Colombia and New York. However, India’s context is distinct due to its massive backlog and reliance on precedent. AI systems trained on historical judgments may perpetuate societal inequalities, such as the overrepresentation of marginalized groups in undertrial prisoners. Legal scholars caution that algorithmic risk assessments could reinforce existing disparities, citing U.S. studies on racial bias in predictive tools. Despite these concerns, the Indian judiciary emphasizes that AI must serve as an aid, not a replacement for human judgment. Mimansa Ambastha, a legal scholar, stated, ‘The moral judgment of the law cannot be delegated to a machine.’

Systemic Challenges and Legal Reforms

India’s case backlog, exceeding 50 million pending cases, poses a systemic threat to its justice system. Over 180,000 cases have remained unresolved for more than three decades, exacerbated by understaffing, procedural delays, and resource shortages. AI is positioned to expedite case management but remains limited to administrative tasks like transcription, legal research, and procedural efficiency. The Kerala High Court mandated Adalat.AI for all subordinate courts in 2025, requiring judges to verify AI outputs and prohibit its use in judgment drafting. This cautious approach reflects a tension between technological efficiency and judicial integrity.

Governance Frameworks and Expert Warnings

Legal reforms in India focus on establishing frameworks to govern AI adoption. The Kerala High Court’s 2025 policy mandates strict oversight, restricting AI to transcription and translation while banning generative AI for decision-making. Judges must verify AI outputs and prevent confidential data uploads to external platforms. The Supreme Court’s proposed rights-based framework assesses institutional readiness, risks, and safeguards such as transparency and accountability. These policies highlight the need for governance to prevent misuse of AI.

Risks of Algorithmic Bias

Experts warn of algorithmic bias and the erosion of human discretion in AI-assisted judicial processes. Studies indicate AI systems trained on historical judgments may inherit societal inequalities, such as the overrepresentation of marginalized groups in undertrial prisoners. Legal scholars note that opaque AI decision-making processes make bias detection and correction difficult, echoing U.S. findings on racial bias in predictive tools. Dr. Ananya Varma, a legal technologist, stated, ‘AI’s opacity creates a black box where errors and biases can go undetected.’ The Indian judiciary’s cautious stance, limiting AI to administrative tasks, aims to preserve judicial discretion.

“'AI’s opacity creates a black box where errors and biases can go undetected.'”

Andhra Pradesh Case Highlights Risks

A 2025 case in Andhra Pradesh underscored risks of AI overreach. A judge cited non-existent AI-generated legal precedents, leading to a Supreme Court ruling of misconduct. The incident prompted investigations into legal professionals and highlighted dangers of relying on unverified AI outputs. Legal scholars argue such cases demonstrate the need for stringent oversight and human verification in AI-assisted processes. Dr. Varma noted, ‘The Andhra Pradesh case serves as a cautionary tale about the perils of algorithmic overreach.’

Balancing Efficiency and Judicial Integrity

India’s judiciary faces balancing efficiency gains from AI with the preservation of judicial integrity. While tools like SUPACE and Adalat.AI address administrative bottlenecks, their integration must avoid undermining the human-centric nature of justice. Legal scholars and policymakers advocate a hybrid model where AI assists with non-adjudicative tasks, allowing judges to focus on core decision-making. The Supreme Court’s proposed rights-based framework emphasizes transparency, accountability, and institutional readiness. Justice Surya Kant stated, ‘AI must be a tool for the judiciary, not a substitute for it.’ Success hinges on balancing innovation with ethical and legal principles.