التزييف العميق والهندسة الاجتماعية 2.0: المخاطر، الفجوات الدفاعية، واستراتيجيات المواجهة الهجينة

Abstract

The cybersecurity landscape is undergoing a radical transformation with the emergence of "Social Engineering 2.0." Attacks no longer rely solely on text manipulation but have expanded to employ generative artificial intelligence techniques, specifically "deepfakes," to create hyper-realistic audio-visual media. This research paper aims to analyze the risks arising from the integration of deepfakes in social engineering attacks, focusing on the current shortcomings of traditional defense mechanisms that fail to detect deepfakes in compressed environments and live communication channels. Through a critical review of recent literature and the proposal of a hybrid defense framework, the study discusses the importance of combining automated technical analysis with human behavioral verification. The findings conclude that technical solutions alone are insufficient in light of the development of competitive generative networks (GANs), recommending the adoption of "zero trust" strategies and out-of-band verification as essential for protecting institutional and societal assets in the age of artificial media.