AI-Powered UPI Fraud Detection and Alert System

Abstract

The AI Powered UPI Fraud Detection and Alert System is a smart mobile application designed to protect users from online financial frauds involving fake brand websites and fraudulent UPI IDs. With the surge in digital payments through Unified Payments Interface (UPI), cybercriminals have begun exploiting user trust by creating fake websites and UPI IDs that mimic genuine businesses or brands. This project provides a real-time fraud detection and alert system that helps users identify suspicious activities before making transactions. The motivation behind this system comes from the growing number of UPI related scams in India, where unsuspecting users are tricked into sending money to fraudulent accounts through fake payment links or cloned websites. Current payment apps lack the ability to verify the authenticity of UPI IDs or detect risky domains. Hence, an intelligent, AI-driven tool is needed to ensure user safety during UPI transactions. The outcome of this project is a fully functional mobile app that verifies UPI IDs. The app is built using Flutter for cross-platform functionality, with Firebase integration for real-time database operations. It uses AI and machine learning models trained to identify patterns in blacklisted UPI IDs, phishing URLs, and risk behaviors. The Firebase collections (blacklisted upi, risk patterns, ml models) help the system store and retrieve fraud data efficiently. The innovation of this project lies in combining AI-based fraud detection, real-time Firebase integration, and a community-driven reporting mechanism, allowing users to contribute to a shared fraud database. Unlike existing payment systems, this application proactively scans and alerts users before transactions occur, significantly reducing the risk of financial scams. The project represents a forward-looking step toward building a secure, intelligent, and trustworthy digital payment ecosystem in India.

Introduction

The text describes an AI-powered UPI Fraud Detection and Alerting System designed to reduce increasing fraud in India’s digital payment ecosystem. With the rapid adoption of UPI, scammers use fake UPI IDs and cloned payment links to deceive users, leading to financial losses. To address this, the proposed system uses machine learning (Random Forest) to detect suspicious UPI IDs in real time by analyzing patterns, anomalies, and historical fraud data.

The system is built using Python for the AI backend, Flutter for a mobile-friendly interface, and Firebase for secure data storage, authentication, and real-time updates. Users can scan or enter UPI IDs and receive instant fraud alerts, while the system also supports community-based reporting to continuously improve detection accuracy. It further includes brand verification to identify fake websites or impersonation attempts.

The literature review highlights that existing fraud detection systems often lack real-time processing, adaptability, explainability, and UPI-specific solutions. Current methods are mostly static, delayed, and not integrated with mobile payment ecosystems.

The proposed methodology includes data preprocessing, ML-based classification, real-time alert generation, and continuous model retraining using new fraud reports. The system architecture consists of a mobile frontend, AI engine, Firebase backend, notification system, admin dashboard, and secure data storage.

Conclusion

This project demonstrates a practical application of AI to reduce UPI financial fraud by monitoring potential fraud before it happens. Using predictive analytics coupled with a user-friendly mobile app, users are warned in real-time against high-risk UPI IDs to prevent monetary loss. Integration of ML models, cloud storage, and a mobile frontend ensures security, efficiency, and real-time fraud detection. The system lays the base for future enhancements such as integrating live transaction systems, adap- tive updating of models, and wider deployment to build trust in UPI transactions.

References

[1] N. P. Khopade & S. M. Vitalkar, “UPI Fraud Detection Using Machine Learning,” International Journal of Research in Interdisciplinary Studies (IJRIS), Vol. 3, No. 6, pp. 24–26, Jun. 2025. [2] Jallapuram Sindhu & Vijaya Sree Swarupa, “UPI Fraud Detection Using Machine Learning Algorithms,” International Journal of Engineering Research and Science & Technology (IJERST), Vol. 20, No. 4, 2024, pp. 57–67. [3] Kothapally Chandini, Akoju Mahender & P. Venkateshwarlu, “UPI Fraud Transaction Detection Using Machine Learning,” International Journal of Engineering Research and Science & Technology (IJERST), Vol. 21, No. 4, 2025, pp. 281–285. [4] Renu Chaudhary, Sakshi Singh, Riddhima Singh, Husain Zaidi & Kanishka Jain, “Fraud Detection in UPI Payments Using Tabular Machine Learning Models,” IJRASET, 2025-10-31. [5] D. Jaya Kumari, G. Tejaswi, N. Jahnavi Nekkanti, A. Korapati, K. Kotakonda & S. Medapati, “AI-Powered UPI Fraud Detection,” International Journal of Innovative Science and Research Technology (IJISRT), Vol. 10, No. 4, 2025, pp. 1208-1213. [6] Akinnagbe O. B. & Akintayo T. A., “The Impact of Machine Learning on Fraud Detection in Digital Payment,” Asian Journal of Science, Technology, Engineering, and Art (AJSTEA), Vol. 3, No. 2, 2025, pp. 191-209. [7] Md Zahin Hossain George, Md Khorshed Alam & Md Tarek Hasan, “Machine Learning for Fraud Detection in Digital Banking: A Systematic Literature Review,” arXiv preprint, Oct. 2025. [8] Junliang Wang, “Fraud Detection in Digital Payment Technologies Using Machine Learning,” Journal of Economic Theory and Business Management, DOI:10.5281/zenodo.10926495.

Copyright

Copyright © 2026 Samruddhi Mahale, Shubhangi More, Anam Shaikh, Bhairavi Patil, Prof. G. V. Barde, Dr. B. S. Shirole. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.