Smarsh unveils AI tool to cut compliance alert noise

Smarsh has launched an AI-based tool designed to filter out low-risk digital communications before they enter compliance review queues, targeting smaller financial firms facing growing volumes of supervisory alerts.

The product, Noise Reduction Agent, is part of Smarsh Professional Archive and aims to reduce false positives in communications surveillance. In early previews, Smarsh reported a more than 60% reduction in false positives, saving more than 40 hours per reviewer per month and increasing redirected capacity by 20%.

Compliance teams across financial services face rising scrutiny and a growing number of channels for client and internal communications. Smaller and mid-sized firms often have similar supervisory obligations to larger institutions but fewer staff dedicated to monitoring and escalation. Many also rely on third-party archiving and supervision tools rather than building systems in-house.

Alert fatigue is a recurring issue in surveillance programmes. Rules-based policies flag keywords, phrases, domains, or message patterns, generating large daily volumes of alerts. Many relate to messages that do not indicate misconduct or regulatory breaches, driving up costs and making it harder to prioritise higher-risk material.

Filtering at ingestion

Noise Reduction Agent applies AI during ingestion within Smarsh Professional Archive, rather than filtering after alerts are generated. It identifies and suppresses content it classifies as low risk, while keeping records available for audit.

Examples of suppressed material include spam, disclaimers, newsletters, promotional text, and automated system messages. Smarsh also cited broader email noise, noting that nearly half of all global email traffic is spam.

The approach reflects a broader shift in compliance technology towards triage and prioritisation to reduce workload created by blanket capture and review requirements. Many regulated firms archive large volumes of communications for recordkeeping and supervision, then apply surveillance policies that can generate high volumes of alerts-many of which are reviewed and cleared without further action.

Smarsh framed the product as a response to pressure on smaller compliance teams. "Compliance teams aren't overwhelmed by misconduct-they're overwhelmed by the volume of noise," said Sheldon Cummings, President, Corporate Business at Smarsh.

"Noise Reduction Agent allows firms to suppress low-risk communications before they ever reach supervision queues, dramatically reducing review backlogs while preserving full regulatory defensibility," Cummings said.

SMB focus

Smarsh is aiming the launch at small and medium-sized businesses, which often operate under similar regulatory obligations as the largest banks while lacking comparable staffing levels. The company did not provide customer names or details of the preview programmes behind the early performance figures.

The product adds to Smarsh's communications governance and archiving portfolio, covering capture, retention, and oversight of digital communications. Smarsh sells into financial services and the public sector, with a client base it says includes large banks, brokerages, insurers, registered investment advisers, and government agencies.

AI has become a prominent feature in compliance and risk technology, as regulators and firms consider how automated tools are governed and tested. Model risk management and governance controls have drawn increased attention as institutions evaluate where AI can be used in supervisory processes and how outputs are validated.

Product positioning

Smarsh said it is extending techniques historically used by large financial institutions into offerings for smaller firms, positioning the tool as part of a move towards autonomous systems that work alongside human reviewers.

"We're proud to deliver a scalable surveillance solution grounded in best-in-class AI governance and model risk management," said Maya Keekattil, VP of Product Management at Smarsh.

"Our mission has been clear: bring enterprise-grade AI supervision to small and mid-sized firms in a way that is accessible, affordable, and built for regulatory confidence," Keekattil said.

Smarsh also pointed to analyst research, saying it was named a Leader for the seventh consecutive time in the 2025 Gartner Magic Quadrant for Digital Communications Governance and Archiving Solutions.

The company said Noise Reduction Agent suppresses low-risk messages while keeping complete records available for audit and regulatory purposes, and plans to roll out the capability within Smarsh Professional Archive for firms seeking to reduce supervision queue volumes.