This project, titled “Zero Trust Architecture-Based Intelligent Cybersecurity and Secure Chip Technologies,” presents a comprehensive cybersecurity initiative spanning cloud infrastructure, AI systems, and hardware-level security. It aims to address Taiwan’s national security priorities through the development and deployment of application-driven technologies across three core domains: Zero Trust Architecture (ZTA), explainable and resilient AI security, and tamper- resistant chip technologies. The project introduces a data-centric ZTA framework optimized for hybrid cloud environments, featuring decentralized identity management, on-premise policy enforcement, and risk-aware access control. In AI security, it implements trapdoor-based defenses to prevent training data leakage in federated learning and builds a multimodal DeepFake detection pipeline for combating disinformation. For hardware security, it develops radiation-hardened Physical Unclonable Function (PUF) chips and pre-silicon electromagnetic (EM) side-channel leakage evaluation tools, enhancing the integrity of chips used in space, defense, and IoT systems. Five interlinked sub-projects will support real-world deployments in Taiwan’s government, military, and semiconductor sectors. The project will also strengthen NTU’s cybersecurity infrastructure and will drive strategic collaborations with leading global research institutes, including Japan’s NICT and France’s CNRS, focusing on federated AI, darknet analysis, privacy-preserving signal processing, and global AI governance. Through these efforts, the project seeks to accelerate the adoption of secure, scalable, and explainable cybersecurity technologies, while reinforcing Taiwan’s digital sovereignty, critical infrastructure protection, and leadership in trusted AI-enabled systems.