Tamper-Resistant Zero Trust Access Management with Intelligent Intrusion Detection

Abstract Modern enterprises increasingly rely on cloud services, mobile devices, and Internet-of- Things (IoT) assets, resulting in highly distributed and dynamic attack surfaces. Traditional perimeter-based security models and static access control schemes are no longer sufficient to protect critical resources against sophisticated adversaries. Zero Trust Architecture (ZTA) has emerged as a leading paradigm that enforces the principle of “never trust, always verify” by continuously evaluating user identity, device posture, and contextual risk 1, 2. However, current ZTA deployments still rely on logically centralized policy decision points and conven- tional logging infrastructures, which raise concerns regarding trust, integrity, and resilience. In parallel, intrusion detection systems (IDS) struggle to keep pace with evolving attack techniques when they depend solely on signature-based or manually tuned rules 14, 15. This paper proposes a blockchain-enabled zero trust access control framework augmented with intelligent, AI-driven intrusion detection for modern cybersecurity systems. The frame- work leverages a permissioned blockchain to record identities, access policies, and audit logs as tamper-evident, verifiable records, thereby decentralizing trust and strengthening accountability 7, 8. Access control decisions are expressed and enforced through smart contracts, while a machine learning–based IDS continuously analyzes network and host telemetry to detect anomalies and high-risk behaviors 17, 18. Detected threats dynami- cally influence access decisions via a risk-adaptive feedback loop. We empirically evaluate the intelligent IDS component using the UNSW-NB15 dataset, a modern benchmark for network intrusion detection research 19. We describe the system and threat models, present the overall architecture, detail the machine learning methodology, and report experimental results on detection performance. A security and architectural discussion highlights how the proposed approach mitigates key attack vectors such as policy manipulation, log tampering, and stealthy lateral movement, il- lustrating the potential of combining blockchain, ZTA, and AI-driven IDS in next-generation cybersecurity systems.