Perception Point, a platform that specializes in providing security to cloud apps, AI-powered email, and web browsers, has introduced an innovative solution to tackle the growing threat of AI-generated email attacks.

By utilizing advanced AI technologies like Large Language Models (LLMs) and Deep Learning, the firm aims to effectively detect and prevent Business Email Compromise (BEC) attacks, which have seen a surge in sophistication with the rise of Generative AI (GenAI).

Hackers Take Advantage of Generative AI

With the rise of AI models, cybercriminals have been leveraging evolving generative AI technology to orchestrate highly targeted attacks against organizations of all sizes. 

The accessibility of GenAI capabilities, which enable the creation of personalized and human-like emails, has empowered these threat actors in social engineering and BEC attacks. 

As spotted by HelpNetSecurity, the alarming statistics from Perception Point’s 2023 Annual Report reveal an 83% growth in BEC attempts, while the DBIR 2023 Report highlights that over 50% of social engineering incidents are attributed to BEC.

Perception Point’s Cutting-Edge Solution

To slow down the BEC attempts, Perception Point has come up with a smart way to deal with the attacks. By launching the LLM-based detection model, recognizing distinctive patterns in LLM-generated text comes to play.

Unlike traditional security vendors relying on contextual and behavioral analysis, Perception Point’s solution utilizes the transformative capabilities of LLMs.

Quicker Way to Detect AI-Generated Email Attacks

Perception Point’s model boasts impressive processing speeds, with an average of 0.06 seconds per email, allowing for near real-time scanning of all email content. This ensures swift and comprehensive analysis. 

Initially trained on an extensive dataset of malicious samples, the model undergoes continuous data training to make it more efficient in detecting the online threats coming on the way.

Three-Phase Architecture

False positives can hurt an AI model even if it’s effective. It can ruin the results and make them appear to be legitimate even when they’re not. 

To decrease the incidents of having false positives, Perception Point delivers a unique 3-phase architecture, per VentureBeat. Following an initial scoring process, the model leverages Transformers and clustering algorithms to categorize email content accurately. 

By integrating additional data such as sender reputation and authentication protocol information, the solution provides comprehensive insights. This meticulous approach enables Perception Point’s solution to accurately determine if an email is AI-generated and assess its potential threat level, ensuring minimal false positives.

“Amid an increasingly complex threat landscape, there is an urgent need for cutting-edge defenses against GenAI-powered threats. We’re being challenged as an industry with yet another avenue that bad actors have come to exploit in their ever-expanding range of attacks,” Perception Point’s CTO Tal Zamir said.

To know more about the company and other things that you can explore about it including blogs, case studies, services, white papers, and more, just visit its official website through this link.

This article first appeared in Tech Times, written by Joseph Henry on June 30, 2023.