Inspiration
Students receive messages every day through WhatsApp, SMS, email, social media, and other platforms. Some contain scams, fake offers, suspicious links, impersonation, or attempts to steal personal information. Students may not always know how to recognize these threats.
I wanted to build a tool that helps students understand whether a message could be dangerous instead of simply telling them "scam" or "not a scam."
What it does
Kavach AI is an AI-powered scam-awareness assistant designed to help students analyze suspicious messages.
A user can provide a message, and Kavach AI analyzes it and provides a fraud-risk score along with understandable reasons explaining why the message may be suspicious.
The goal is education as well as detection: students can learn what warning signs to look for, such as suspicious links, urgency, requests for sensitive information, impersonation, or unusual payment requests.
How I built it
I built the project using HTML, CSS, and JavaScript for the interface, with AI-powered analysis for message classification and explanation. The project is designed around explainable AI so that the user can understand the reasoning behind the result.
I also focused on keeping the interface simple enough for students to use without needing cybersecurity knowledge.
Challenges
One challenge was making the system useful without simply producing a yes/no answer. A useful safety tool needs to explain the warning signs clearly.
Another challenge was thinking about responsible AI: the result should be treated as an assessment rather than an absolute guarantee. Users should still verify suspicious messages through trusted sources.
What I learned
This project helped me learn more about AI integration, prompt design, web development, explainable AI, and responsible use of AI for cybersecurity awareness.
Building Kavach AI also taught me that technology can be more useful when it helps people understand a problem rather than simply giving them an answer.
Log in or sign up for Devpost to join the conversation.