Smarsh has published a new report examining how artificial intelligence can reduce the volume of low-value communications requiring manual review in financial institutions. Based on a two-month evaluation involving 15 financial institutions and more than 2.1 million emails, the report explores the practical application of AI in communications compliance workflows.

According to the report, AI-driven filtering reduced manual email review volumes by up to 70% while achieving a classification accuracy of 99.65% for low-value communications enabling compliance teams to redirect significant supervisory capacity from routine reviews to higher-risk activities, including investigations and escalations.

Rather than replacing existing compliance controls, the report argues that AI can strengthen supervisory processes by reducing operational burden while maintaining auditability and regulatory defensibility. It also highlights the growing importance of scalable supervision as financial institutions face increasing volumes of digital communications without corresponding increases in compliance resources.

Overall, the report provides a practical case study of how AI can improve the efficiency of communications surveillance in regulated financial services while allowing compliance teams to focus more effectively on genuine risk.