Law, technology and child protection: A contemporary analysis of ai in cases of children in conflict with law

Abstract

Artificial intelligence is increasingly influencing policing, criminal investigation, judicial administration, risk assessment, digital evidence analysis and rehabilitation across criminal justice systems. Its application to children in conflict with law, however, presents qualitatively different legal and ethical concerns because juvenile justice is founded not merely upon determination of responsibility but upon protection, rehabilitation, reintegration, dignity and the possibility of a fresh start. In India, the Juvenile Justice (Care and Protection of Children) Act, 2015 establishes a specialised child-centred framework governed by principles including presumption of innocence, dignity and worth, participation, best interests, equality and non-discrimination, privacy and confidentiality, institutionalisation as a measure of last resort, diversion, fresh start and natural justice. These principles impose stringent normative limitations upon the use of artificial intelligence in juvenile justice. AI technologies may potentially assist authorities in analysing digital evidence, managing case pendency, identifying children requiring services, facilitating translation and accessibility, improving rehabilitation planning and detecting systemic patterns. At the same time, predictive policing, facial recognition, automated risk scoring, profiling and algorithm-assisted assessments may reproduce historical social inequalities, create opaque classifications and convert socioeconomic disadvantage into apparently scientific predictions of future criminality.

The concern becomes particularly important in relation to children aged sixteen to eighteen alleged to have committed heinous offences. Section 15 of the Juvenile Justice Act requires the Juvenile Justice Board to make an individualised preliminary assessment concerning the child's mental and physical capacity, understanding of consequences and circumstances surrounding the alleged offence. The Supreme Court's decision in Barun Chandra Thakur v. Master Bholu emphasises the delicacy and complexity of this process. AI cannot legitimately replace the Board, psychologists, psycho-social workers or other qualified experts in making such an assessment.

The article analyses artificial intelligence through India's constitutional framework, juvenile justice legislation, privacy jurisprudence, the Digital Personal Data Protection Act, 2023, and the India AI Governance Guidelines released in 2025. As of August 2026, India follows a principle-based, sectoral and techno-legal AI governance model rather than a comprehensive horizontal AI statute. The national guidelines expressly emphasise human oversight, fairness, accountability and proportionate safeguards for sensitive and high-risk AI applications. The article compares this approach with international child-rights standards and the European Union AI Act. It concludes that AI should remain a supportive and auditable instrument in juvenile justice and should never become an autonomous mechanism for determining culpability, dangerousness, transfer to adult trial, detention or rehabilitation outcomes. A rights-based framework of child-specific algorithmic impact assessment, meaningful human oversight, explainability, data minimisation, independent audits and effective remedies is essential if technology is to function as a protective rather than punitive instrument.