Scam Shield

An AI-powered web app that helps seniors identify scams from messages and screenshots, providing clear explanations, risk scores, and practical safety advice in simple language.

Comment

ScamShield AI

Inspiration

The idea for ScamShield AI came from seeing how often seniors receive fake bank alerts, lottery messages, and phishing scams. Many older adults struggle to identify these threats and often don't have someone nearby to ask for help. We wanted to build a simple AI-powered tool that explains suspicious messages in plain language and helps users stay safe online.

What it does

ScamShield AI allows users to paste suspicious messages or upload screenshots for AI analysis. It provides a risk score, identifies warning signs, explains the scam in simple language, and recommends safe next steps.

How we built it

We started by using Lovable to quickly generate the initial project structure and user interface. From there, we customized and extended the application to meet our goals.

We integrated Fireworks AI (Kimi K2.6) to power intelligent scam detection for both text messages and screenshots. We implemented server-side AI functions, connected the frontend with the backend, refined prompts for accurate analysis, and added fallback heuristic detection for reliability.

The frontend was built with React, TanStack Start, TypeScript, and Tailwind CSS, while Git and GitHub were used for version control and collaboration. Throughout development, we tested, debugged, and refined the application to improve accuracy, usability, and performance.

Tech Stack: Lovable, React, TanStack Start, TypeScript, Tailwind CSS, Fireworks AI (Kimi K2.6), Git, GitHub.

Challenges

Designing an AI response that is accurate, easy to understand, and suitable for senior citizens while keeping the interface simple and accessible.

What's next

  • More scam categories
  • Multi-language support
  • Voice assistance
  • Real-time scam database
  • Mobile application

Impact

While many AI assistants can answer questions, ScamShield AI focuses on solving a real cybersecurity problem. Unlike generic AI chatbots, our platform is purpose-built to detect scams from both messages and screenshots, explain why something is suspicious in simple language, and provide actionable steps users can take to protect themselves. By combining AI with rule-based verification, ScamShield AI delivers more reliable, practical, and user-friendly scam detection.

Built With

Share this project:

Updates

Submission history