Shield-AI

• 

  Shield-AI Dashboard on Mobile"

Inspiration Online scams and digital fraud are rising in Pakistan. Many people in my community lose money because they can't identify fake messages. I wanted to build a simple AI tool that anyone can use on their mobile to stay safe.

What it does Shield-AI uses the Gemini 1.5 Pro model to analyze suspicious SMS, emails, and links. It tells the user if a message is a scam and provides a safety rating. It also helps identify AI-generated deepfake images.

How we built it I built this entire project on my mobile phone using Google AI Studio. I used advanced Prompt Engineering to ensure the AI focuses on cybersecurity patterns and local scam tactics.

Challenges we ran into The biggest challenge was working without a laptop. Managing the AI Studio interface and testing prompts on a small screen was difficult, but I stayed focused on the goal.

Accomplishments that we're proud of

I am incredibly proud of building Shield-AI entirely on my mobile phone, as I do not have a laptop. Successfully configuring the Gemini 1.5 Pro model to accurately detect local scam patterns and creating a functional AI tool despite technical limitations is my biggest achievement so far."

What we learned I learned how to use the Gemini API and how to write system instructions that make AI a specialized security expert.

What's next for Shield-AI

"The next step for Shield-AI is to develop a dedicated mobile application that can scan incoming SMS in real-time. I also plan to add support for local languages (like Urdu) so that non-English speakers can stay safe from digital fraud. Eventually, I want to integrate an automated reporting system that alerts local authorities about common scam patterns."

Built With

• ai
• api
• engineering
• gemini
• google
• prompt
• studio