Artificial intelligence in criminal justice: legal challenges to fair trial guarantees and procedural safeguards

Abstract

The increasing use of artificial intelligence in criminal justice is no longer a distant prospect but an emerging reality that is gradually reshaping procedural practices. Algorithmic tools are increasingly involved in risk assessment, data analysis, and evidentiary evaluation, particularly in cases involving complex digital environments. The main aim of this study is to examine the legal implications of integrating artificial intelligence into criminal proceedings, with a particular focus on its impact on fair trial guarantees and procedural safeguards. The analysis draws on international human rights standards, selected jurisprudence of the European Court of Human Rights, and recent developments in European Union regulatory approaches to artificial intelligence. The research is based on a doctrinal and comparative legal methodology, allowing for an assessment of how existing legal frameworks respond to the challenges posed by algorithmic decision-making. Particular attention is given to issues of transparency, contestability, and the ability of parties to effectively challenge AI-generated outcomes. The findings indicate that the opacity of algorithmic systems creates significant difficulties for maintaining equality of arms, particularly in situations where procedural rules do not adequately ensure meaningful access to the reasoning underlying automated outputs. It is argued that without clearly defined safeguards ensuring transparency, accountability, and proportionality, the use of artificial intelligence risks undermining the fairness and legitimacy of criminal proceedings