Anti-scam Ear Web App

Imagine your phone warning you before a scammer convinces you. Our web app helps people detect and prevent suspicious or fraudulent calls in real time.

Comment

Inspiration

Inspiration This project was inspired by a personal experience. I once received a phone call that sounded legitimate and urgent, and for a moment, I almost fell for it. That experience made me realize how easily people can be manipulated through suspicious or fraudulent calls,especially when fear, urgency, or authority is involved. I wanted to build something that could help people pause, stay calm, and make safer decisions before it’s too late.

What it does

The web app helps users identify and avoid suspicious or dubious phone calls. It works by analyzing call-related inputs and risk signals to determine whether a call may be fraudulent. When a potential threat is detected, the app alerts the user and provides guidance, helping them stay informed and protected from call-based scams and manipulation. The goal is simple: reduce phone call fraud and give users peace of mind.

How we built it

The project was built as a web application with a focus on simplicity and speed. We integrated AI capabilities using the Gemini API to help analyze patterns and indicators that could suggest suspicious behavior. The frontend was designed to be user-friendly, while the backend handles data processing and API communication efficiently.

Challenges we ran into

One of the major challenges was integrating the Gemini API into the web application. Setting up the connection, handling API responses correctly, and managing errors took significant time and experimentation. There were also challenges around structuring prompts effectively to get accurate and useful results from the AI. Despite these hurdles, overcoming them helped us gain a deeper understanding of AI integration and web app architecture.

Accomplishments that we're proud of

Even without a live API key today, we successfully built a fully functional backend and frontend system that demonstrates how AI can be integrated for scam detection. The mock response system allows judges to test the application immediately, while the backend is fully ready to work with a real Gemini API key once it is obtained. We are proud of the clean architecture, robust error handling, and the way the project demonstrates real-world AI integration.

What we learned

Through building the Anti-Scam Ear Web App, we learned how to integrate AI APIs into a web application securely and efficiently. We gained experience handling environment variables, managing backend-frontend communication, and designing a system that works even without immediate access to a live API key. We also learned the importance of modular design and providing fallback solutions for testing and demos.

What's next for Anti-scam Ear

• The next step is to obtain a Gemini API key, which will allow the backend to fully process AI-generated responses instead of using mock data. Once the key is obtained, the system will provide real-time scam detection for user input. We also plan to improve the frontend for better usability, explore additional AI models to increase accuracy, and make the web app fully deployable so it can help users identify potential scams in real time.

Built With

  • css
  • firebase-(firestore-&-authentication)
  • gemini-3-ai-api
  • google-identity-api
  • google-sign-in-(google-identity-services)
  • html
  • javascript
Share this project:

Updates