The rapid advancement of Industrial Control Systems (ICS) in critical infrastructures has elevated their vulnerability to sophisticated cyber threats. Traditional security mechanisms often fall short in addressing the dynamic and complex nature of modern cyber-attacks. This paper presents an Autonomous AI-Based Cyber Defense framework designed to enhance the resilience and sustainability of ICS. Leveraging real-time monitoring, predictive analytics, and autonomous response mechanisms, the proposed model mitigates risks with minimal human intervention, ensuring uninterrupted operations and long-term system sustainability. Key components include anomaly detection, threat prediction using machine learning models, and adaptive response mechanisms that dynamically adjust security policies based on evolving threat landscapes. The implementation of this intelligent defense strategy demonstrates significant improvements in threat detection accuracy, incident response time, and system robustness, fostering sustainable industrial automation in an era of escalating cyber risks.