A Comprehensive Usage of AI in Prevention of Social Engineering Attack

2025

Closed access Metadata (CSL-JSON)
DOI 10.1109/csnt64827.2025.10968143
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

AI has become a tool for social engineering attackers, enhancing the likelihood of successful attacks. However, as the saying goes, "Diamonds cut diamonds." In this research, we have enlisted the usage of AI to initiate, launch, counter and prevent social engineering attacks. We have explored not only traditional forms of social engineering attacks with AI, such as phishing, eavesdropping, shoulder surfing, dumpster diving, piggybacking, and tailgating but also the latest forms of AI-enabled social engineering attacks, known as vishing, baiting, Smishing, Whaling, Quishing and Quid Pro Quo. This research paper also conducted a systematic literature review of AI-based defending techniques like AI Sentry,