Islamabad, November 1, 2025 – The Federal Board of Revenue (FBR) has revealed that cutting-edge machine learning technology is now playing a pivotal role in identifying high-risk cases for tax audits, transforming the traditional approach to compliance monitoring in Pakistan.
According to an official report, the FBR has implemented a Machine Learning–based Compliance Risk Management (CRM) system across all major field formations. The system is designed to detect anomalies, assess, rank, and address tax compliance risks using diverse datasets and advanced analytical models.
Unlike conventional audit selection methods, the CRM system leverages machine learning algorithms such as clustering and multivariate analysis to identify hidden patterns and irregularities within massive datasets. These models help flag potential cases of tax evasion and fraud that might otherwise go unnoticed.
The FBR stated that the new system also supports targeted awareness campaigns, behavioral interventions, and incentive mechanisms to improve voluntary compliance. Additionally, built-in feedback features ensure that taxpayers’ concerns are addressed more efficiently.
The CRM dashboard is now operational across large, medium, and corporate tax offices, providing officers with real-time access to risk ratings and profiling of taxpayers.
This data-driven approach allows the FBR to optimize its audit resources, focusing efforts on high-risk cases and ultimately enhancing revenue collection while promoting transparency and accountability in Pakistan’s tax administration.