Mechanisms for Building Trust in AI-Assisted Justice Systems Through Cognitive, Emotional and Socio-Psychological Factors

artificial intelligence (AI) is progressively gaining traction within legal and judicial spheres, which necessitates a re-examination of concepts such as transparency, fairness, reliability, and legitimacy. The implementation of explainable artificial intelligence (XAI) is a key element in ensuring trust in such systems, though ‘explainability’ does not in itself guarantee that people will begin to trust these systems. Purpose: the study aims to provide a detailed analysis and systematization of available scientific information on psychological mechanisms underlying the formation of trust in explainable artificial intelligence systems applied in the field of justice, with the cognitive, emotional, and socio-psychological aspects considered. Methods: the study employed methods of analysis and synthesis to explore the complex system of the formation of trust in explainable artificial intelligence in the field of justice. An analysis of a wide range of open sources, including scientific articles, legal documents and research reports, made it possible to categorize facts regarding the perception of computerized judicial decisions and the level of trust in them. Results: the research has established that trust in AI applied in the field of justice develops at the intersection of three key aspects: the understanding of the fundamentals of AI operations, emotional acceptance of the technology, and its adherence to social norms. The author explains the significance of transparency and explainability of AI decisions, of the ability of AI systems to interact with users on an emotional level, and of the role of public opinion and expert examination in trust development. The study has found correlations between perceived accuracy of AI systems and trust levels, elucidated the importance of empathetic characteristics in AI interfaces and the influence of group dynamics on trust formation. Conclusions: the findings demonstrate the need for a comprehensive approach to AI implementation in the legal sector that would consider both psychological and technological factors. The paper provides practical recommendations for increasing trust in AI in justice, including the development of training programs, standardized explainability schemes, and transparent audit mechanisms. Recommended areas of future research include studies into cross-cultural variations in attitudes to AI, long-term strategic consequences and ramifications of justice automation, and the development of methods for assessing the effectiveness of human-AI collaboration in court.

Перевод пока недоступен