Artificial Intelligence and Law in India: Promise, Peril, and the Limits of the Algorithm
Author: Mritika Raj
Student, KES JAYANTILAL H PATEL LAW COLLEGE
—————————————————————————
3 Quick Takeaways
- Artificial intelligence offers significant advantages to India’s criminal justice system, including faster analysis of forensic and digital evidence, smarter legal research, and greater administrative efficiency, but these benefits come with serious risks of algorithmic bias, opacity, and threats to the right to a fair trial.
- The “black box” problem in AI, where internal decision-making processes are not transparent or explainable, poses a direct challenge to the principles of evidence law, including the requirements of reliability, authenticity, and the right of cross-examination.
- India’s existing legal framework, including the Section 65B certification requirement affirmed in Anvar P.V. v. P.K. Basheer (2014), provides a starting point for regulating AI-generated evidence, but a dedicated, dynamic statutory framework for AI in criminal justice is urgently needed.
Introduction
Artificial intelligence has become an inseparable part of daily life. What once took entire days, months, or even years can now be accomplished in seconds. Fields as varied as information technology, research, forensic science, art, and literature have all integrated AI into their workflows. On a personal level, people use AI for career decisions, projects, and entertainment.
But can we trust AI when it comes to the legal field?
This article examines the role of AI in India’s criminal justice system, with a particular focus on evidence law. It explores both the advantages and the legal and ethical challenges that arise when AI tools are deployed in law enforcement, prosecution, and judicial decision-making. It asks: to what extent can AI be used without compromising the accuracy, fairness, and integrity of the legal system?
AI’s intellectual journey began long before the current boom. While its rapid growth accelerated in the 2020s, its roots stretch back to the 1940s and 1950s. John McCarthy, widely known as the father of AI, coined the term and organised the 1956 Dartmouth Workshop that launched artificial intelligence as a formal field. Earlier contributions came from Alan Turing’s foundational Turing Test (1950), Warren McCulloch and Walter Pitts’s model of artificial neurons (1943), and the work of Marvin Minsky, Nathaniel Rochester, and Claude Shannon. What we use today is the product of nearly a century of collective intellectual effort.
The Bright Side: Where AI Helps
When it comes to processing large volumes of data, AI surpasses human capacity by a significant margin. In the legal field, it can rapidly analyse forensic data, geographic crime data, medical evidence, and digital records. For lawyers and judges, AI dramatically accelerates legal research, including the identification of case laws, statutory provisions, and precedents from vast databases. Tasks that previously consumed hours can now be completed in minutes.
Beyond research, AI has improved administrative efficiency across courts and legal institutions. It aids in document drafting, case summarisation, and even translation, making the justice system more accessible to non-English-speaking litigants in a country where the judiciary primarily operates in English. AI-driven tools for predictive policing can identify crime hotspots and patterns, potentially allowing prevention before incidents escalate. Facial recognition and automated video analytics have similarly supported law enforcement investigations.
The Government of India has acknowledged this potential, allocating Rs. 7,210 crores for the e-Courts Phase 3 project, with Rs. 53.57 crore specifically earmarked for the integration of AI and blockchain technologies across High Courts. This signals a strong institutional commitment to the digital transformation of justice.
The Dark Side: Where AI Falls Short
Justice is not an algorithm. It is something felt, not computed. A machine operating on commands and code cannot fully account for the complexity of human circumstances: the context of an offender’s upbringing, the pressures of their environment, or the mitigating factors that courts have long recognised as relevant to sentencing and liability.
The most serious technical problem is the “black box” phenomenon: the lack of transparency and explainability in AI systems whose internal logic is not visible or understandable to the people using them. When applied to facial recognition, predictive policing, risk assessment, and forensic tools such as DNA analysis or voice recognition, black box AI creates significant dangers for criminal justice. If a system produces an output that recommends detention or inculpates an accused, but no one can explain how that conclusion was reached, the right to challenge evidence is effectively undermined.
Algorithmic bias presents an equally serious concern. AI systems trained on historical data carry the prejudices embedded in that data. In criminal justice, where marginalised communities are already disproportionately represented in enforcement data, an AI system that learns from that data will likely reproduce and amplify existing inequalities, producing discriminatory outcomes that violate the constitutional guarantee of equality before the law.
The risks of deepfakes and fabricated digital evidence further complicate matters. As AI tools grow more sophisticated, so does the potential for manufactured content to be presented as authentic evidence in court proceedings.
The Legal Framework: What Exists and What Is Missing
India’s traditional evidentiary framework under the Indian Evidence Act, 1872, requires that evidence be genuine, reliable, and subject to cross-examination. AI-generated evidence challenges each of these requirements.
Indian courts have begun engaging with these questions through the lens of digital evidence law. In Anvar P.V. v. P.K. Basheer, (2014) 10 SCC 473, the Supreme Court held that electronic records are admissible only if accompanied by a certificate under Section 65B, establishing authentication and verification as non-negotiable conditions for digital evidence. This ruling is directly relevant to AI-generated outputs: without verifiable certification of how such evidence was produced, it cannot safely enter the judicial process.
However, beyond Section 65B, India currently lacks a dedicated, dynamic statutory framework for AI-generated evidence. There are no binding guidelines governing the admissibility, reliability testing, or disclosure requirements for outputs produced by predictive policing systems, facial recognition tools, or forensic AI algorithms. While empirical reliance on such tools by law enforcement is growing, judicial and statutory guidance has not kept pace.
Fundamental rights must not be the casualty of technological enthusiasm. The presumption of innocence, the right to a fair trial, and the right of both parties to be heard cannot be compromised in any framework that accommodates AI.
Conclusion
AI in the criminal justice system is neither the answer nor the enemy. It is a tool, and like all tools, its value depends entirely on how carefully and thoughtfully it is used.
India must resist the temptation to adopt AI wholesale without first establishing the legal infrastructure to govern it. Courses on AI literacy should begin in schools, creating a generation that understands both its promise and its risks. Most urgently, Parliament must enact a comprehensive framework that sets clear standards for the use, verification, and admissibility of AI-generated evidence in criminal proceedings.
The concern about a new digital divide is real and pressing. In a country where access to justice is already unequal, allowing AI to further stratify the system would be a constitutional failure. From the richest to the poorest citizen, justice must remain equally available. That principle cannot be algorithmic.
Disclaimer: The views expressed in this article are those of the author and do not necessarily reflect the views of The Lawscape.
The Lawscape — clear, practical legal insight for students and future lawyers.