This study examines the use of AI-based fraud detection systems among financial institutions in Iran, the United Arab Emirates, and Qatar, with a particular focus on trust, transparency, and perceived fairness. Despite the promise of AI operations in identifying financial anomalies, uncertain decision-making processes, and algorithmic bias limit their widespread adoption, especially in regulated banking sectors. This study uses a quantitative strategy based on partial least squares structural equation modeling and multi-group analysis of survey responses from four hundred banking professionals, such as auditors and compliance officers. The study shows that transparency greatly enhances trust, which is the main predictor of AI adoption. Perceived fairness moderates the negative effects of algorithmic bias, emphasizing its important role in building system credibility. Subgroup analysis reveals distinct regional and professional variations in trust and fairness sensitivity, with internal auditors and those highly exposed to AI showing greater readiness for adoption. Regulatory compliance also emerges as a positive factor in adoption. This research identifies transparent, explainable, and fairness-sensitive AI tools as essential to promote adoption in regulated sectors. The findings provide guidance for promoting responsible and trust-based AI implementations in fraud detection.