Factors Impacting Successful Prosecution of Cases in Courts: Content Analyses of Judgments Using Large Language Models

• In Event: Digital Crime and Decentralized Finance: Artificial Intelligence, Cryptocurrency, and Large Language Models

Abstract

The study examines court judgements of cases registered and investigated by the Anti-Corruption Bureau, mandated to prevent and detect corruption by government officers. This research using AI-powered Large Language Models (LLMs) conducts content analysis of the judgements and quantifies the various characteristics of the facts detected during the investigation of the case. Data pertaining to variables of the nature of the offense, the characteristics of the offender, legal charges against the accused, the quality of the prosecution and the defense, an assessment of the probability of guilt, the procedural justice safeguards extended to the accused, and the outcomes of the case are compiled. The study assesses the factors that lead to a successful prosecution or defense of a case. The research also evaluates the validity and reliability of the content analyses done using LLMs by manually examining a sample of the judgements analyzed by using AI tools.

Authors

• Vinod Kumar TK, University of North Carolina Wilmington

• Arvind Verma, Indiana University Bloomington