4 key takeaways from Red Sift Radar’s live briefing

If you missed the introduction of our latest innovation, Red Sift Radar, don’t worry – we’ve got you covered! Here are the 5 key takeaways from our launch webinar with Red Sift’s CEO Rahul Powar, Senior Director of Sales Engineering Billy McDiarmid and Zachary Bennett, IT & Security Manager at FHC.

Watch the live briefing below or keep scrolling for the written recap.

But first, a quick recap. What’s Radar?

Red Sift Radar is the upskilled LLM that leverages internet-scale cybersecurity intelligence and the terminal skills of an analyst to help security teams identify and resolve issues up to 10x faster. Radar launched with a first-of-its-kind DMARC integration through Red Sift OnDMARC, with plans for broader integration across the Red Sift portfolio in the coming months.

A new LLM innovation from a company with roots in AI

The Radar launch briefing kicked off with Rahul discussing Red Sift’s deep experience with LLMs and the innovation journey leading to Radar.

In July 2020, the Red Sift team first saw an LLM in action when they gained early beta access to GPT-3. Rahul noted, “This early insight allowed us to start ideating. […] In our swimlanes, we care about making cybersecurity accessible, actionable, and simplified where possible […] so how do we use this technology to try and do that?”.

This initial insight paved the way for a series of milestones: launching the first GPT-4-powered feature in Brand Trust, rolling out AI-powered filtering in Certificates, and eventually prototyping the first version of Radar as a plugin in the OpenAI store. However, the team decided to bring Radar in-house to ensure customer data security and seamless integration with Red Sift’s platform.

“If it’s not there at your point of need, all the time, it becomes a lot less useful. We wanted to bring the AI to where your problem actually is.”

Rahul Powar, CEO & Cofounder, Red Sift

How Red Sift overcame the limitations of traditional LLMs

Rahul expanded on how Red Sift addressed key limitations of traditional LLMs to build a best-in-class cybersecurity solution:

  1. Isolation: Traditional LLMs require users to know exactly what questions to ask, and they exist as standalone tools. Red Sift solved this by integrating Radar directly into OnDMARC, offering ‘jumping-off points’ where users may have complex questions, allowing Radar to assist right when it’s needed.
  2. Incomplete Data: LLMs are often isolated from real-time data. Red Sift tackled this by integrating Radar with its extensive data sets and proprietary tools, exposing Radar to public data, open SSL tooling, and insights from global points of presence. This ensures Radar can provide comprehensive, real-time insights.
  3. Inconsistency: To avoid variable outputs, Red Sift created “guardrails,” or playbooks, that help Radar think through multi-step security queries reliably and consistently. To read more about playbooks, visit the Radar launch blog.

Radar empowers resource-strapped security teams 

Red Sift Radar enhances the security posture of organizations by flagging misconfigurations and exposures to teams and helping them remediate them on the fly, without the need for additional headcount. By combining AI with deep industry knowledge, Radar’s integration with OnDMARC helps security teams:

  1. Find and fix misconfigured records: Radar provides market-leading identification of errors in SPF and DMARC records, ensuring email authentication settings are correctly aligned.
  2. Quickly identify unknown senders: Radar enables fast evaluation of unknown IP addresses to identify and classify potential threats.
  3. Better understand DMARC failure reports: Unlock better analysis of DMARC failure reports to differentiate between legitimate forwarders and email spoofing attempts.

“In the first 24 hours of use, Radar saved us $4,000 by identifying a fraudulent invoice from its header information.”

Zachary Bennett, IT & Security Manager at FHC

In addition to its in-app interactions inside OnDMARC, Billy showed off the Email Analyzer tool that saved Zachary $4000 by spotting a fraudulent invoice by its header information. After a suspicious email was flagged by FHC’s AP/AR team, Zachary used Radar to investigate the email. “Normally, I would have had to look at the header and check in Microsoft where this email was coming from,” he said. “Radar made a complete report with everything I needed to understand that the email was spoofed. It came from an externally compromised email which Radar confirmed with an IP address check, told us that the username did not match the current company’s email address, and revealed that the domain had been registered 2 days prior… Ingenious!”

Best of all, Radar’s chat interface allows users to ask questions in natural language, making it accessible even to those without deep technical expertise. As Billy McDiarmid noted, “We want to allow people with all levels of experience in cybersecurity to use our software … and provide a simplified platform that shares insight and data continuously and automatically across all of our applications.”

Radar delivers real-world impact

Zachary went on to share how Radar transformed his approach to cybersecurity, turning him from an LLM skeptic to a believer. In addition to invoice fraud detection, Radar helped FHC consolidate its cybersecurity tools, boost the team’s efficiency, and streamline day-to-day operations. 

“Radar [has] replaced the need for alternate tools so instead of having a bunch of browser tabs open like MXToolBox and DNS Checkers, it’s all within one application.” He added, “My team isn’t ginormous. We have a lot of different jobs. Having everything in one place without moving between different tools means you have time to do other tasks. The integration with the Red Sift OnDMARC is a huge selling point […] because it’s all there, on the side panel.”

“Radar goes way beyond email security […] it’s an all-in-one tool.”

Zachary Bennett, IT & Security Manager at FHC

To read more about how Radar delivered value to our beta customers, visit the launch blog here.


Red Sift Radar is now available to existing Red Sift customers across our OnDMARC, Brand Trust, and Certificates product portfolio. If you missed the webinar, you can catch up with the recording on YouTube

Ready to get started? Request your account today. 

PUBLISHED BY

Francesca Rünger-Field

11 Oct. 2024

SHARE ARTICLE:

Categories

AI

Recent Posts

VIEW ALL
DMARC

Over 60% of healthcare organizations remain unprotected against data breaches

Sean Costigan

Introduction Red Sift’s analysis of healthcare organizations that reported large breaches to the Department of Health & Human Services (HHS) in 2023-2024 uncovered a troubling trend: post-breach, 61% remain unprotected against phishing and domain spoofing due to weak or nonexistent DMARC policies. DMARC (Domain-based Message Authentication, Reporting & Conformance) is a widely recognized security…

Read more
Awards

Red Sift wins 2025 Cybersecurity Excellence Award for OnDMARC

Jack Lilley

Executive Summary: Red Sift OnDMARC has been recognized with the 2025 Cybersecurity Excellence Award for its advanced email security solutions. By leveraging AI-powered tools like Red Sift Radar for security issues and Dynamic DNS Guardian for real-time monitoring, OnDMARC provides businesses with robust protection against phishing, spoofing, and business email compromise (BEC).  Key takeaways:…

Read more
Product Release

Red Sift’s Winter ‘24/’25 Quarterly Product Release

Francesca Rünger-Field

This quarter, we’re making security faster, smarter, and more proactive with updates that improve threat detection, reduce manual work, and prevent threats before they escalate. Highlights include: Brand Trust  Executive Impersonation: Detect unauthorized use of leadership identities By uploading and managing executive images in Brand Trust, security teams can detect and monitor unauthorized use…

Read more
AI

Enhanced logo detection with AI: A hybrid approach

Phong Nguyen

Executive Summary: Accurate logo detection is essential for protecting brands against misuse and fraudulent activities. Red Sift’s hybrid AI approach enhances detection precision, effectively balancing the reduction of false positives with the identification of genuine threats. This article: Introduction Logo detection is crucial for brand protection, helping identify logo misuse in lookalike domains and fraudulent…

Read more