Cofense, the leading provider of intelligence-driven phishing defense solutions, is unveiling new AI-powered capabilities for its Phishing Detection and Response (PDR) platform that work to ease SOC analyst workloads with AI spam reduction. SOC analysts, plagued by heavy workloads and alert fatigue, benefit from the ability to sort through alert noise and focus on genuine threats with Cofense’s new AI functionality.
Today’s SOC teams are burdened with a high volume of alerts—many of which are false alarms—atop their already extensive list of security tasks, according to Cofense. As phishing attacks continue to evolve—with email being one of the largest security threat vectors for enterprises—being able to stay ahead of security threats despite overwhelming workloads is crucial.
“Spam is a costly distraction for SOCs, constituting at least 30% of emails that bypass SEGs and other filters,” explained Jason Reinard, senior vice president, product engineering, Cofense. “This overwhelming volume of spam creates a significant burden on SOC analysts, hindering their ability to focus on critical security analysis.”
In response, Cofense’s new AI-powered spam reduction utilizes Bayesian Machine Learning (ML) to understand SOC environments and then identify and filter out spam automatically. This dramatically enhances SOC workflows by enabling them to train the ML model to process out the unique spam overwhelming their inboxes, creating a more targeted approach to security analysis, according to Reinard.
Cofense’s AI-based spam filter “learns [SOC environments] based on training that is directed by the SOC analysts’ classification efforts,” said Reinard. “Emails that the SOC have classified as spam train the model to identify spammy tokens, whereas emails classified as legitimate business communications weigh the other side of the model. This has proven to be an effective means to identify what is considered spam within each unique organization.”
“As new emails come in that are not identified by the ML as being spam, analysts can selectively train the model on those misses. Once the model is trained, and automation is defined, the scores can be leveraged to set up Triggers to process out the spam,” Reinard continued.
Additionally, by keeping emails local and never requiring exportation to external cloud data lakes, the Cofense PDR platform—unlike other AI security products—offers true data privacy, drives compliance, and supports best-practice security methods, according to the company.
To learn more about Cofense’s new AI-powered spam filtering, please visit https://cofense.com/.