Leveraging AI and ML Applications for Robust EV Information Security : A Review

Abstract

  Although integrating artificial intelligence and machine learning into electric vehicles to improve cybersecurity is gaining in popularity, the practical applications of this concept are still relatively unexplored. In this review paper, we summarize the current status of AI and ML-driven security in electric vehicles (EV) discussing their responsibilities on authentication, intrusion detection as well attack prevention. Using a deatiled literature review the analysis finds that indeed, we are finding more and more applications such as the above utilizing ML techniques but also deep learning, neural networks. This paper sheds light on the growing interest to link blockchain technology with AI and ML for better EV security. This joint solution will directly address the evolving automotive threat landscape as vehicles become more connected and autonomous where system complexity, its interconnectedness with other systems mandates advanced security regime. Advantages: About 75% of research explores intrusion detection; ~20%, authentication and the remaining 5% considers attack preventionDeep learning features as an independent method lead with majority researchers followed by neural networks, according to this study. This Spiking Adoption Of AI/ML In EVs Clearly Seeks For Continuous Lines of Research to Tackle Potential Threat Permutations. EVs will be increasingly autonomous and connected, making them prime targets for malicious actors who could choose to target the vehicles or their users—and therefore preemptive security enforcement is necessary. As such, security methods for EVs in the future could go from reactive to predictive with an AI-powered frame that can perceive and relieve risks before they create. And this proactivity will be mandatory to target the complexity of upcoming EV systems, and firmly establish their vulnerability towards cyber threats.