ANALYSIS OF ZERO TRUST-BASED DATA SECURITY ARCHITECTURE DESIGN FOR CYBER THREAT MITIGATION

Abstract

The radical shift in global digital threat dynamics, predicted to reach a critical point in 2026, demands a transformation of the cybersecurity paradigm from a static perimeter model to a more adaptive architecture. This research aims to design a data security application model that integrates Zero Trust Architecture (ZTA) with an Artificial Intelligence (AI)-based analytical engine to comprehensively protect critical information assets, both at rest, in transit, and in use. The urgency of this study is driven by the ineffectiveness of conventional security in dealing with synthetic identity attacks, autonomous malware, and the challenges of quantum computing against classical cryptography. Although ZTA adoption has grown rapidly globally, referring to the NIST SP 800-207 standard, there is still a significant empirical gap in validating ZTA's effectiveness against Generative AI (GenAI) attacks. The design methodology in this research carries the principle of "never trust, always verify" by integrating Machine Learning to automate security policies based on User and Entity Behavior Analytics (UEBA). The results of this design are expected to be not only reactive, but also predictive in mitigating security risks in the midst of cloud ecosystems, IoT, and hybrid work systems.