Physical layer security is an emerging approach to strengthen traditional encryption methods in response to growing cybersecurity threats. These threats and vulnerabilities disrupt the normal operation of networks, emphasizing the need for new security solutions. Among the various types of attacks, jamming is a major threat at the physical layer, rendering vehicular communications particularly vulnerable. Numerous jamming detection techniques have been proposed over the past decade, but many have proven ineffective. Furthermore, traditional intrusion detection techniques have limitations in detecting jamming attacks, underscoring the necessity for more specialized and improved solutions. This paper investigates the effectiveness of machine learning models in detecting jamming attacks. We trained, evaluated, and tested machine learning algorithms specifically Random Forests and Support Vector Machines. Simulation results demonstrate that the Random Forest-based algorithm detects jammers with high accuracy while reducing computational complexity, making it a promising solution for fast and precise jamming detection.