Automation of virtual inspections in fiscal auditing: technological design for the UEA

This study designs, implements, and evaluates an integrated automation system for virtual statutory audit inspections, incorporating efficiency accelerators and emerging technologies. It addresses critical weaknesses in manual document verification through architectures based on artificial intelligence, natural language processing, and blockchain to ensure data integrity. A quantitative descriptive-correlational methodology was applied to 25 statutory auditors inspected by the Special Administrative Unit over the past two years. The system integrates the eight components of ISQM 1: risk assessment, governance and leadership, ethical requirements, client acceptance and continuance, engagement performance, resource allocation, information and communication, and monitoring and remediation. Results show a 73% reduction in processing time, accuracy improvement from 71% to 94.3%, and a 67% decrease in manual workload. The architecture includes four modules: automated document classification (85% faster categorization), consistency analysis engine (94.3% inconsistency detection), real-time management dashboard, and preventive alert system (78% reduction in formal findings). Perception analysis indicates 80% support automation and 100% expect 60–70% time savings, confirming alignment with international quality standards and establishing a precedent for fiscal supervision modernization in Colombia