The rapid advancement of autonomous driving technology has transformed modern transportation, offering enhanced safety, efficiency, and convenience. However, as these vehicles become increasingly connected and reliant on complex software and sensor-based systems, they also become prime targets for a wide range of cyber and privacy threats. This review paper comprehensively examines the current landscape of security and privacy in autonomous driving systems. We explore emerging attack vectors targeting key components such as sensor perception, vehicle-to-everything (V2X) communication, machine learning models, and internal control systems. Particular attention is given to adversarial machine learning, GPS spoofing, Controller Area Network (CAN) bus attacks, and data privacy breaches. In parallel, we evaluate existing defense mechanisms and mitigation strategies, including intrusion detection systems (IDS), secure communication protocols, hardware-based security modules, and privacy-preserving architectures. We also highlight key challenges in securing autonomous systems, identify gaps in current research, and propose directions for future work to build resilient and trustworthy autonomous vehicles. This review aims to provide researchers and practitioners with a consolidated foundation for understanding and advancing the security posture of next-generation autonomous driving technologies.