Across the connected vehicles ecosystem, cybersecurity has emerged as a critical challenge. Modern vehicles are more than mechanical systems; they are intricate networks of software, data, and advanced technologies that are highly connected and thus vulnerable to cyberattacks. This raises the stakes for robust and forward-thinking cybersecurity strategies. Enter Artificial Intelligence (AI), a cutting-edge solution with immense potential to elevate automotive cybersecurity systems.
Automotive cybersecurity ranks high on the list of mandatory compliances. Security assessment is, therefore, needed at all levels including devices / components, ECUs, interfaces and at entire vehicle level. The industry needs a continual and standardized methodology for security assessment and identification of must-fix security issues at earlier phases. There is also a critical need for maintaining and tracking security assessment information for product compliances and certifications.
Automotive systems have seen tremendous transformation after the advent of sensor-based real-time communication. With several features being introduced in a vehicle, there is ever increasing sensor-based real-time in-vehicle communication. For the connected vehicle there are communications received from outside vehicles via 5G, Wi-Fi, BT, OBD etc. The vehicles are becoming more software oriented and various safety systems, real-time communication and user experience technologies are controlled by software.
All of this becomes even more critical with the advent of SDV architecture where HPC-based distributed clusters of ECUs are communicating with the central hub, remote software updates have to take place with secure OTA infrastructure having vehicle telemetry support. Need to ensure security at every instance of communication, storage, boot-up etc., is critical. There is ever increasing demand for shortening time required to identify critical assets, respective threats, risks, vulnerabilities and mitigations, convert mitigations into system and software level requirements and create / link test cases for each requirement.
This paper unpacks how AI can play a pivotal role in five key areas of automotive cybersecurity, how AI-led solutions can be developed, and why collaboration among stakeholders is critical to creating robust, reliable, and adaptive security frameworks.