Paradigm Shift in Mobile Authentication Against Generative AI Attacks: A Real-Time HPV Framework

Abstract

The rapid advancement of generative AI technologies has introduced new threats such as deepfakes, voice cloning, and synthetic behavioral patterns, thereby undermining the reliability of existing mobile authentication systems.Traditional mobile authentication has primarily focused on identifying registered users; however, in generative AI environments, distinguishing real humans from AI-generated synthetic identities has emerged as a critical security challenge.Through a convergent approach combining mobile security, biometric recognition, wearable technology, and generative AI detection, this study analyzes the limitations of existing mobile authentication methods against generative AI attacks.It proposes a real-time Human Presence Verification (HPV)-based authentication framework.The proposed framework integrates mobile devices with wearable and connected devices to continuously verify biometric signals, including heart rate, motion, and behavioral continuity, to verify human presence in real time.In addition, the proposed framework enables the detection of spoofing attempts by leveraging the asynchrony between AI-generated synthetic signals and authentic human biometric responses.This study comparatively analyzes conventional authentication methods and the proposed framework in terms of resilience against generative AI threats, security robustness, and continuous authentication capability.The results suggest that real-time HPV can overcome the limitations of conventional static biometric authentication and provide a new paradigm for mobile authentication in the era of generative AI.