An Intelligent Zero Trust Architecture Model for Mitigating Authentication Threats and Vulnerabilities in Cloud-Based Services

The widespread adoption of Cloud-Based Services has significantly increased the surface area for cyber threats, particularly targeting authentication mechanisms, which remain among the most vulnerable components of cloud security. This study aimed to address these challenges by developing and evaluating an Intelligent Zero Trust Architecture model tailored to mitigate authentication-related threats in Cloud-Based Services environments. Data was sourced from public repositories, including Kaggle and the National Institute for Standards and Technology MITRE Corporation’s Adversarial Tactics, Techniques, & Common Knowledge (ATT&CK) framework. The study utilized two trust signals: Behavioral targeting system users and Contextual targeting system devices. Based on the trust signals, two machine learning models—Keystroke Dynamics and Device Location—were developed using Binary Logistic Regression, achieving a combined average accuracy of 80.63%, with a residual ineffectiveness rate of 19.37%. The Intelligent Zero-Trust Architecture Threat Mitigation Model was introduced to reclassify threat severity scores, resulting in the downgrading of all authentication threats to Low Severity, demonstrating a mitigation effectiveness exceeding 80%. This research contributes to the field of cybersecurity by presenting a validated, intelligent, and context-aware Intelligent Zero-Trust Architecture model capable of enhancing identity and access management in dynamic cloud environments. The findings offer actionable insights for cloud architects, cybersecurity professionals, and policymakers aiming to strengthen trust, reduce attack surfaces, and improve threat resilience across digital infrastructure.