Malware is a serious and constantly evolving threat to cybersecurity, affecting individuals, businesses, and governments worldwide. Studies indicate that over 90% of business enterprises have experienced malware-related issues in recent years. Traditional malware detection methods, such as signature-based approaches, often fail to detect novel threats like zero-day attacks, new variations of known threats, and insider threats. With the rapid advancements in artificial intelligence (AI) and machine learning (ML), more effective malware detection systems can be developed. Kavach Shakti, an AI-driven malware detection system, aims to serve as a shield against malware by utilizing deep learning models, including Decision Trees, Neural Networks, and Random Forests. These AI algorithms aid in the efficient and accurate identification of threats, reducing false positives and improving detection rates. Kavach Shakti integrates static and dynamic analysis, machine learning models, and behavioral profiling to enhance malware detection accuracy and enable real-time threat monitoring. The Random Forest and Gradient Boosting models demonstrated high malware detection accuracy of 99.51% demonstrating the potential of ensemble learning methods in cybersecurity defenses.