From Reactive to Proactive: An AI-Enabled Human-Centered Framework for Telecommunication Fraud Prevention

Telecommunication fraud has become an increasingly adaptive form of communication-mediated deception, further intensified by recent advances in artificial intelligence (AI). Existing anti-fraud efforts have made progress in detection, warning, and general risk education, but less attention has been given to how AI can support users’ protective capacities during fraud-related interactions. This paper addresses this gap by developing an AI-enabled human-centered framework for telecom fraud prevention from a human augmentation perspective. Based on a conceptual synthesis of research on fraud vulnerability and prevention approaches, the framework organizes fraud vulnerability across exposure opportunities, individual susceptibility, situated decision-making conditions, social-relational contexts, and fraud-type differences. It then maps these mechanisms onto three augmentation pathways, including sensory augmentation for situational awareness and risk cue perception, cognitive augmentation for reflection, verification, and self-regulation, and social augmentation for trusted consultation and timely help. The framework further translates these pathways into design functions, applicable contexts, evaluation indicators, and ethical safeguards. By linking fraud vulnerability with AI-enabled human augmentation, the study reframes proactive telecom fraud prevention as a capability-oriented design problem that combines technical detection with human judgment, social support, and responsible governance.