Artificial Intelligence and Blockchain in Financial Reporting
0 viewsAbstract
Financial reporting functions are moving towards more automation, artificial intelligence (AI), and distributed-ledger technologies. This paper aims to explore whether faster and better reporting is achieved through an AI–block chain converged reporting environment when compared to a traditional manual baseline in terms of reporting speed, transaction accuracy, audit efficiency, fraud-detection accuracy, and reporting cost. The underlying dataset is the course-provided comparative case study of a medium-sized e- commerce company consisting of approximately 10,000 transactions over a three-month period. Two reporting processes are compared – a traditional versus technology-enabled near real-time transaction ingestion with AI- based classification and anomaly detection and shared permissioned block chain ledger. Monthly closing time decreased by 5.5 days to 1.5 days, transaction accuracy improved by reducing error rates to 0.9% from 3.8%, quarterly audit hours decreased by 55 hours to 65, monthly reporting cost declined by $7,000 to $11,000. Fraud-detection accuracy improved from 62% accuracy to 91%. Due to underlying transaction- level dataset not being provided, results are descriptive in nature and not intended to be claimed as independently statistically validated. The paper concludes with lessons learned and future work in terms of implementation cost, governance and skills, and regulatory uncertainty.
Keywords
Agency Theory, Artificial Intelligence, Audit Efficiency, Blockchain Technology, Financial Reporting, Fraud Detection, Robotic Process Automation, Smart Contracts, Transaction Accuracy