Examining the Legal Recognition of AI-Generated Evidence in Civil Proceedings

ABSTRACT
The use of AI in commercial, financial and interpersonal communication has made evidentiary issues in civil litigation more complicated. Traditional legal frameworks, which were designed for assessing static documents and human testimonies, are hard to process when evaluating machine-generated dynamic outputs, deepfake synthesis, predictive algorithmic reports and data compilations. The present research paper critically analyzes the legal status and admissibility of AI-generated evidence in civil proceedings, particularly the provisions of Bharatiya Sakshya Adhiniyam, 2023 (BSA) and the provisions of the Indian Evidence Act, 1872 (IEA). The study addresses the procedural issues of authentication, hearsay exceptions, opacity of algorithms (the "black box" problem), and the chain of custody. The paper examines whether the existing statutory certifications (e.g., Section 63 of the BSA, which is known by the name Section 65B of the IEA) are adequate to provide validation for machine-generated evidence without human validation. Finally, the paper suggests a special judicial procedure, new legislation that incorporates specific authentication requirements, and judicial panels of algorithmic experts to guarantee procedural fairness and protect the integrity of the evidence in civil cases.