Synergizing Edge AI and Quantum Machine Learning for Real-Time Cyber Threat Mitigation
Shashank Solanki, Rituraj Sinha · 2025
The escalation of the complexity of cyber threats must be countered by traditional signature- and rule-based security approaches. In this study, we propose a hybrid Edge AI–Quantum Machine Learning (QML) framework that employs variational quantum circuits and classical neural networks towards real-time per–device threat detection. Using three case studies, we validate the framework: (1) fraud detection in high frequency trading with 17% more true positives and 22% less false positives; (2) inference times under 100 ms for IoT anomaly detection; and (3) reduction of over 25% in deepfake misclassification. The built system is built end-to-end with an open-source stack. Finally, regulatory and ethical considerations (GDPR, data, privacy, international cybersecurity protocols, etc., Budapest Convention) are discussed. In presenting this work, we present a scalable and adaptive model for next-generation cybersecurity.