Inspiration Cloud storage platforms typically focus on authentication at login but provide limited protection against insider threats and post-authentication attacks. We wanted to build a solution that not only verifies user identity but also continuously monitors behavior to prevent potential data leaks.

What it does SecureVault AI is a secure cloud vault that combines triple-layer authentication with AI-driven behavioral monitoring.

Authentication includes:

  • Password verification
  • Voice biometric authentication
  • SMS OTP verification After login, the AI Security Guard analyzes user behavior, detects suspicious activities such as rapid downloads, sensitive file access, and unusual activity patterns, and automatically restricts access when risk thresholds are exceeded.

How we built it The system was developed using Python, Flask, SQLite, HTML, CSS, and JavaScript. Voice authentication is powered by OpenAI Whisper and SpeechBrain ECAPA-TDNN. Behavioral monitoring uses custom risk-scoring techniques based on activity patterns, filename analysis, and anomaly detection. SMS OTP delivery is implemented through MacroDroid webhook integration.

Challenges we ran into The main challenges were integrating multiple authentication mechanisms into a seamless workflow, tuning behavioral detection thresholds to reduce false alarms, and ensuring reliable voice authentication and SMS OTP delivery.

Accomplishments that we're proud of

  • Developed a complete AI-powered secure cloud vault
  • Implemented triple-layer authentication
  • Built an AI behavioral monitoring engine
  • Enabled predictive data leak prevention
  • Created a practical cybersecurity solution using open-source technologies

What we learned We gained hands-on experience in cybersecurity, biometric authentication, machine learning integration, behavioral analytics, and secure application development.

What's next for AI-Based Multi-Factor Authentication with Predictive Data Leak Prevention for Secure Cloud Vault Future improvements include advanced anomaly detection models, continuous voice authentication, cloud deployment, mobile integration, and enhanced security analytics dashboards.

Built With

Share this project:

Updates

posted an update

SecureVault: AI-Based Multi-Factor Authentication System with Predictive Data Leak Prevention has officially been submitted to Summer Hackathon 2026.

Our project combines voice biometrics, SMS OTP verification, and AI-driven behavioral monitoring to secure cloud storage against unauthorized access and insider threats. Over the coming days, we will continue refining the user experience, improving threat detection capabilities, and preparing our final demonstration.

Stay tuned for updates and feedback.

Log in or sign up for Devpost to join the conversation.