Current technological developments have led to a great deal of embedded sensors, connected objects, and their related networks and communication to be present in the transport area involving Autonomous Vehicles (AVs), aircraft, trains, as well as road infrastructures. Various types of signals and connections, occurring on the Internet of Vehicle (IoV), are vulnerable to security attacks, which can cause the system to fail with serious consequences on the user’s safety. Research on IoV security focuses on securing communication between nodes. Only a few studies have investigated the relationships between security and safety in IoV. Our approach addresses this gap by providing semantic-based analysis to jointly explore safety and security. We propose a semantic-based Artificial Intelligence (AI) reasoning tool to analyse the causal relationships between cyber security and safety for IoV and AVs. This tool runs on the ontology, named Security-Safety Internet of Vehicles (SSIoV) ontology, which represents both security-safety knowledge about IoV domain. Our goal is to perform reasoning and inferences on security vulnerabilities and their impact on safety risks, based on data extracted from real-world scenarios. This research falls in the research areas of cyber-security, because: (a) it involves securing current and future vehicles and charging infrastructures; (b) it uses a semantic AI approach for enhancing cyber-defence; (c) it detects IoV and AV components, vulnerabilities, and risks. Therefore, this tool is also useful to improve preventive cyber-defence capabilities in the IoV and AVs area. Finally, this study contributes to enhance the safety of Switzerland’s IoV-critical road infrastructure.