Inspiration

Online scams, phishing websites, and fake news are spreading faster than ever. In 2024, nearly 3 million phishing incidents in Malaysia caused over RM1.5 billion in losses. Many users think they can spot fake sites, but scammers are getting increasingly sophisticated. We wanted to create a tool that helps users instantly verify the trustworthiness of any website or link before they click.

What it does

RealCheck AI evaluates any website or link in real-time and produces a risk score with a clear explanation. It analyzes domain information, SSL certificates, URL structure, textual patterns, and cross-checks known scam databases. Users can see at a glance whether a website is safe or suspicious, helping them avoid scams and misinformation.

How we built it

Collected datasets of known phishing URLs, scam reports, and legitimate websites. Trained an AI classifier using features like domain age, SSL validity, URL structure, and content patterns. Built a web interface where users paste links and instantly get a risk score with an explanation.

Challenges we ran into

The biggest challenge was evolving phishing tactics — scammers constantly change domains and site structures. Another challenge was making the AI output interpretable, so users trust the risk score and understand why a site is flagged. Balancing accuracy with explainability took several iterations.

Accomplishments that we're proud of

  • Developed a working prototype that can evaluate links in real-time.
  • Successfully combined multiple signals into a single, interpretable risk score.
  • Built a user-friendly interface that even non-technical users can use confidently.

What we learned

We learned that AI can meaningfully improve online safety, but user trust depends on clear explanations, not just a score. Combining multiple signals—metadata, content, reputation—is key to detecting suspicious websites. We also learned the importance of rapid iteration to keep up with evolving scams.

What's next for RealCheck AI

Expand to browser extensions and mobile apps for real-time protection. Integrate with email and social media platforms to detect suspicious links automatically. Continuously update our database and model to adapt to new phishing techniques. Explore regional expansion beyond Malaysia, tackling scams across Southeast Asia.

Built With

Share this project:

Updates