This study examines the synergy between e-fiscalization and Artificial Intelligence (AI) as a transformative framework for modern tax administration. While e-fiscalization enables real-time digital reporting of transactions and generates large volumes of structured data, AI provides the analytical capacity to process and interpret it effectively. The research explores how integrating these technologies enhances financial transparency, strengthens compliance, and reduces tax evasion, particularly through advanced techniques such as machine learning, network analysis, and data mining. The paper adopts a mixed-methods approach, combining comparative analysis of tax performance indicators before and after the implementation of e-fiscalization with a case study focused on the Albanian tax system. Secondary data from institutional reports and international organizations are utilized to assess improvements in efficiency and risk detection. Findings suggest that AI-driven systems significantly improve the targeting of tax audits, enable real-time identification of fraudulent transactions such as fictitious invoicing schemes, and enhance taxpayer services through automated assistance tools. However, the study also identifies critical challenges, including data quality limitations, algorithmic transparency concerns, and cybersecurity risks. These issues highlight the necessity for robust regulatory frameworks and ethical guidelines governing AI deployment in public administration. The research concludes that the integration of AI into e-fiscalization systems represents a strategic necessity rather than a technological option, positioning tax administrations to move from reactive enforcement toward proactive, data-driven governance.
CANI et al. (Fri,) studied this question.