Biography
Dr. D. Jayanthi is an Associate Professor in the Department of Information Technology at Sri Venkateswara College of Engineering (SVCE), Sriperumbudur, Tamil Nadu, with over 22 years of teaching and research experience. She holds a Ph.D. from Anna University, Chennai, and serves as the Institutional Academic Coordinator.
Her research interests include Artificial Intelligence, Machine Learning, Deep Learning, Internet of Things (IoT), Cloud Computing, and intelligent healthcare and weather forecasting systems. She has published research papers in reputed international journals, including Elsevier, Wiley, and MDPI publications, and has received international recognition for her research on COVID-19 forecasting using deep learning, including an appreciation certificate from Elsevier.
Dr. Jayanthi has successfully completed an ISRO-funded research project on deep learning-based weather forecasting and received an institutional seed grant for brain tumour detection and classification using hybrid deep learning models and IoT. Her research contributions also include industry-sponsored projects, patents, and interdisciplinary applications of AI. She has received best paper awards at national and international conferences and has served as a resource person in various AICTE ATAL FDPs, workshops, guest lectures and hands on training in other institutions, keynote contributor, reviewer in Elsevier and Springer Journals, and session chair for various academic and professional events.
Her academic contributions span research, teaching, innovation, and institutional development, with a strong focus on applying emerging technologies to address real-world challenges.
Abstract
Title: Trust No Face, Believe No Voice: The New Frontier of AI-Enabled Social Engineering
Abstract: The rapid advancement of generative artificial intelligence has transformed social engineering through deepfake videos, cloned voices, and highly personalized deceptive messages. Traditional detection systems often focus on identifying manipulated media, overlooking the broader context of identity impersonation, psychological manipulation, and fraudulent intent. It is important to identify a novel, context-aware approach to detecting AI-enabled social engineering by integrating multimodal authenticity analysis, identity consistency, manipulation patterns, and action-based risk assessment. The CAVE (Context-Aware Verification and Escalation) framework aims to identify suspicious interactions, provide explainable risk assessments, and enable timely intervention, even when individual media components appear authentic. The system integrates specialized audio-visual and textual encoders with a cross-modal consistency module and an interpretable risk-fusion mechanism. An uncertainty-aware decision layer is designed to distinguish low-risk communications, suspicious interactions, and high-risk requests requiring independent verification. To address the limitations of conventional evaluation, the framework will be assessed using controlled multimodal attack scenarios, unseen generation techniques, and communication distortions. The intended contribution is a shift from isolated media authenticity classification toward context-sensitive detection of deceptive interactions and actionable risk. The system identifies whether integrating identity, intent, and action-level evidence improves robustness and supports more transparent cybersecurity decisions under evolving AI-enabled social engineering threats.
