A Privacy-Aware Input Validation and Data Masking Framework for Secure AI-Based Web Applications: A Practical Implementation in DataWhiz

This paper presents a privacy-aware framework for secure AI-based web applications, implemented in the DataWhiz platform using the MERN stack (MongoDB, Express.js, React.js, Node.js). The framework integrates input validation, sensitive data detection, and selective masking mechanisms to improve privacy protection in web-based data analysis systems. The proposed system automatically identifies potentially sensitive information such as email addresses, phone numbers, and IP addresses using rule-based detection and pattern matching techniques. It also applies input sanitization and validation to reduce risks associated with malicious or unsafe user inputs. The framework is evaluated through practical implementation and testing within the DataWhiz platform. Results indicate that the system provides effective privacy-aware protection while maintaining usability and acceptable performance for real-world web applications. This work demonstrates that practical privacy and security mechanisms can be integrated into modern full-stack web applications without requiring complex infrastructure, providing a foundation for future research in privacy-preserving collaborative systems and secure data analysis platforms.