Natural Justice In The Age Of Artificial Intelligence: Reassessing Audi Alteram Partem In Automated Administrative Decision-Making

Abstract

Indian administrative law recognises audi alteram partem as an important component of the constitutional guarantee of fairness under Articles 14 and 21. The principle is based on the expectation that an affected person should be able to understand and respond to the material on which a decision is made. The increasing use of artificial intelligence in administrative decision-making raises concerns about this assumption, particularly where the working of an algorithm is not readily accessible. Adopting a doctrinal and comparative approach, this article examines three forms of AI involvement in decision-making and considers the Bombay High Court’s decision in KMG Wires Pvt. Ltd. v. National Faceless Assessment Centre (2025 ) and the Supreme Court’s decision in Pooja Ramesh Singh v. Jammu & Kashmir Bank Ltd. (2026) , alongside selected developments from the EU, Netherlands and United States. The article argues that the key concern is whether affected persons can meaningfully challenge decisions when relevant inputs and reasoning remain inaccessible. It proposes a limited shift in the burden of establishing prejudice where algorithmic opacity prevents such prejudice from being fairly assessed.