This study proposes a real-time Intrusion Detection System (IDS) to enhance maritime cybersecurity against increasingly sophisticated threats, particularly ransomware attacks, in the era of digital transformation in the shipping industry. The proposed system integrates Support Vector Machine (SVM), a machine learning technique, with AI-based future prediction to detect ransomware intrusions early and respond in real time by analyzing onboard network traffic, system logs, file access records, and user activity data. Experiments were conducted using real operational data from three 24,000 TEU-class container vessels. The system demonstrated an average detection accuracy of approximately 63%, showing a 15% improvement over conventional methods. The false positive rate remained below 3%, and the average response time was measured to be under 0.5 seconds, confirming the effectiveness of the system in promptly mitigating ransomware spread. Designed as a lightweight system suitable for shipboard environments, the proposed IDS also offers potential for integration with autonomous vessel operations and smart port security architectures. This paper presents a practical solution to maritime cybersecurity threats and provides a foundation for future development of integrated security frameworks capable of countering diverse cyberattacks.