Data leakage is one of the most significant threats to data security and must be addressed to reduce its risks. It occurs when sensitive information is inadvertently disclosed to unauthorized parties, exposing it to cyber attackers and criminals who can use the data to commit identity theft, steal credit card details, or sell the data on the dark web. Therefore, this paper focuses on effectively using the Zero Trust principle to solve the data leakage problem using well-established and widely used algorithms. It assumes that any user or device, regardless of network location, poses a potential threat. Thus, providing a better way to protect user privacy in a way that accurately perceives leakage is designed to significantly improve the reliability of our results and effectively address them, offer reliable security guarantees for network security, and achieve continuous monitoring. By adopting a Zero Trust approach, organizations can significantly improve their data security posture and protect sensitive information in an increasingly hostile digital environment. The experimental results demonstrate the effectiveness of our strategy, achieving an accuracy of 96.41% and confirming its robustness in real-world applications.