Scamshield AI

💡 Inspiration

Online scams targeting students and job seekers have become increasingly common. Many fake job offers and internship scams look professional, create urgency, and pressure users to pay fees or share sensitive information. We were inspired by real stories of students losing money simply because they didn’t know how to verify whether an offer was legitimate.

We wanted to build something that doesn’t just detect scams, but actually helps people make safer decisions in moments of uncertainty.

🛡️ What It Does

ScamShield AI is an AI-powered scam detection tool that analyzes job postings, emails, and suspicious messages to identify scam indicators.

It provides:

A clear Trust Score (0–100) indicating risk level
Simple explanations of why something is suspicious
Community scam evidence from public discussions when available
Recommended next steps so users know what to do immediately

Instead of overwhelming users with technical details, ScamShield AI focuses on clarity, transparency, and guidance.

⚙️ How We Built It

ScamShield AI combines multiple layers of detection:

Rule-based scam signals (payment requests, urgency language, unrealistic offers)
AI-powered text analysis for scam likelihood
Domain and link safety checks
Community-driven evidence using publicly available scam reports

The system is designed to work both as a Chrome extension (for in-context scanning) and a web dashboard (for detailed reports), making scam detection accessible right where users browse.

🧠 What We Learned

Explainability matters as much as accuracy when dealing with security tools
Users trust systems more when they understand why a decision was made
Combining AI with real-world community warnings creates stronger confidence
Ethical design and safety disclaimers are essential in cybersecurity tools

This project helped us think deeply about responsible AI and user-centric security design.

🚧 Challenges We Faced

Balancing detection accuracy with clear, non-alarmist explanations
Ensuring the system works reliably even without external API access
Presenting complex scam signals in a simple, readable way
Designing a demo-friendly experience within a short hackathon timeframe

Each challenge pushed us to simplify and focus on what truly helps users.

🚀 What’s Next

With more time, we would:

Add WhatsApp and SMS scam detection
Support regional languages for wider accessibility
Build a community reporting and verification system
Expand scam education with interactive tips and alerts

Note: ScamShield AI provides guidance, not guarantees. Community discussions are public and may be unverified. Users should independently verify critical decisions.