FUSING DATA FROM VARIOUS LAYERS (NETWORK, ENDPOINT, CLOUD) TO DETECT NEVER-BEFORE-SEEN THREATS USING AI

Abstract

Due to the growing sophistication of cyber threats, organisations need to deploy smart 
systems that can identify attacks that have never been witnessed before. The paper proposes 
research into integrating network, endpoint, and cloud data with the help of AI to increase the 
effectiveness of “never-before-seen” threat detection and response. The study compares the 
conventional strategies with those based on multi-layered AI models through case studies, 
experimental findings, and literature to demonstrate their superiority to the classic approaches in 
the field of federated learning and anomaly detection. Important challenges, the implications 
regarding practice, and future directions are also discussed to inform the implementation of secure 
and scalable implementation. These results further underline the importance of adaptable 
cybersecurity that is done in conjunction with the help of AI. 

,