Perception Point has developed an artificial intelligence model that uses large language models (LLMs) and deep learning architecture to detect and prevent business email compromise (BEC) attacks.
The new AI model comes as cybercriminals are increasingly using generative AI technologies to launch sophisticated and targeted BEC attacks, Perception Point said in a prepared statement.
What Generative AI Means for Cybercriminals and BEC Attacks
Generative AI is easily accessible to cybercriminals. Meanwhile, BEC attacks are being "supercharged" by generative AI, Perception Point stated.
Cybercriminals can use generative AI to reduce the time it takes to perform target research and reconnaissance for attacks. They can complete these tasks and many others in minutes. As such, generative AI empowers cybercriminals to "work faster, and on a much larger scale than ever before," Perception Point said.
How Perception Point's New AI Model Works
Perception Point's model leverages Transformers, which are AI models that utilize technologies similar to those used in OpenAI ChatGPT and Google Bard and are capable of understanding the semantic context of text, the company noted.
Transformers help Perception Point's model identify unique patterns in LLM-generated text, which is paramount to detecting generative AI-based threats, the company stated.
Furthermore, Perception Point's model processes incoming emails at an average speed of 0.06 seconds. It also is aligned with Perception Point's ability to scan 100% of content in near real time.
Perception Point's model has been trained on hundreds of thousands of malicious samples caught by the company. It also is continuously updated with new data to maximize its effectiveness, the company indicated.
Perception Point's AI Model Minimizes the Risk of False Positives
Perception Point's model uses a three-phase architecture to reduce the risk of false-positive email security alerts, the company stated.
First, the model scores an email-based threat. It then categorizes the content using Transformers and clustering algorithms, integrating insights from these steps with additional data. From here, the model can predict whether an email is AI-generated and if it potentially poses a threat, Perception Point said.
Tal Zamir, Perception Point's chief technology officer, commented on his company's new AI model and how it can help organizations protect against BEC attacks:
"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. By reversing this dynamic and proactively leveraging AI for detection, we are able to prevent these threats before they even reach the user's inbox – a paradigm shift in the fight against BEC attacks."