Inspiration

Online scams are becoming harder to recognize. We wanted to create a simple tool that helps people check suspicious content before they click, reply, share information, or send money.

What it does

ScamShield AI analyzes suspicious messages, emails, links, screenshots, voice calls, QR codes, phone numbers, and UPI IDs. It provides a risk score, red flags, evidence, and recommended safety actions.

How we built it

We built ScamShield AI using React, TypeScript, and Vite, with Google Gemini for AI analysis and Firebase for authentication and user data. The project is hosted through Google AI Studio.

Challenges we ran into

Our main challenges were handling different input types, improving AI consistency, reducing false warnings, securing user data, and keeping the interface simple and responsive.

Accomplishments that we're proud of

We created one platform that can analyze multiple types of scams and explain not only whether something is suspicious, but also why it may be dangerous.

What we learned

We learned that scam detection depends heavily on context, not just keywords. We also gained experience with Gemini, Firebase, authentication, AI APIs, and frontend development.

What's next for ScamShield AI

We plan to improve detection accuracy, multilingual support, scam-pattern recognition, payment-scam analysis, and overall security and reliability.

Built With

Share this project:

Updates