Explainable AI in Automated Financial Reconciliation: A Survey-Based Analysis Across Multi-Location Enterprises
TL;DR
Respondents perceive explainability as a meaningful contributor to reconciliation quality, financial control, and decision support, though cross-location consistency and the clarity of AI-generated explanations remain comparatively weaker areas that warrant targeted investment.
Abstract
Background: Multi-location enterprises increasingly rely on artificial intelligence to reconcile financial transactions across branches, banking systems, and enterprise resource planning platforms, yet the opacity of many AI models continues to undermine the confidence auditors, controllers, and regulators place in automated outputs (Cerneviciene & Kabašinskas, 2024). Explainable AI (XAI) has been proposed as a bridge between automation and accountability, though empirical evidence on its organizational value in reconciliation contexts specifically — as opposed to credit scoring or fraud detection, where most XAI-finance research has concentrated — remains limited. Methods: A cross-sectional, structured online survey was administered to 175 professionals working in finance, accounting, auditing, information technology, data analytics, and organizational management. Seven constructs — Explainable AI Capability, AI Explainability and Transparency, Automated Reconciliation Quality, Exception Detection Capability, Multi-Location Consistency, Financial Control Effectiveness, and Audit and Decision Support — were measured on five-point Likert scales and analyzed descriptively, via Pearson correlation, and via exploratory factor analysis (principal component extraction). Results: All constructs were rated favorably (means 4.09–4.24). Automated Reconciliation Quality was rated highest (M = 4.24); Multi-Location Consistency was rated lowest (M = 4.09). Automation Quality correlated most strongly with Reconciliation Effectiveness (r = .748), followed by XAI Capability (r = .703). Five factors explained 86.90% of total variance. Conclusion: Respondents perceive explainability as a meaningful contributor to reconciliation quality, financial control, and decision support, though cross-location consistency and the clarity of AI-generated explanations remain comparatively weaker areas that warrant targeted investment.
