Inspiration
As digital technologies have developed quickly, we noticed that a lot of students and young people have been tricked into clicking on phishing links, fake job applications, and suspicious sites. They seem highly credible, which makes it hard for users who are not experienced enough to recognize them. That is why our team came up with an idea of creating a tool that will serve as a guide for all those users. It should be easy to use and based on artificial intelligence technology.
What it does
TrustLens AI is an intelligent web-based application that helps users detect and avoid online scams by analyzing suspicious messages, emails, or links in real time. The user simply inputs the content they want to verify, and the system uses AI to evaluate it. It then provides a structured output that includes a risk classification (Low, Medium, or High), a clear explanation of the potential threat, and actionable recommendations for the user. By combining advanced AI analysis with a simple and user-friendly interface, the platform enables users especially students and non-technical individuals—to make informed decisions and stay protected in the digital environment.
How we built it
The project is built as a basic online app with three primary components: Frontend (HTML, CSS, JavaScript) Features a form where users paste suspicious content Displays results from the analysis in clear terms Backend (Node.js / Python) Interacts with the users and works with the Gemini API Processes the output of the AI Gemini API (AI Core) Analyses the given content through Natural Language Processing Detects scamming activity and behavior through pattern recognition Outputs response which includes information about the risk level and warning. Workflow User Input → Frontend → Backend → Gemini API → Response → Display to User
Challenges we ran into
In creating the concept and the system architecture, some difficulties were encountered: Creating effective prompts: To ensure that Gemini will be able to provide consistent outputs Maintaining the balance between usability and power: So that the system can be user-friendly yet intelligent at the same time Dealing with ambiguous input data: Certain messages may not be explicitly classified as either safe or dangerous. Establishing user confidence: So that they will believe and understand the results provided by the system.
What we learned
Throughout this project, we have acquired significant insight into many aspects of cybersecurity, such as:
- Cybersecurity threats in the real world, including phishing and social engineering techniques.
- The workings of AI models like Gemini in processing human languages.
- Creating an intuitive and easy-to-use web interface for non-tech-savvy users.
- Creating systems that consist of front-end, back-end, and API interactions.
- Prompting an AI to give useful outputs.
- It is essential to have clarity and explainability in outputs generated by AI models.
What's next for TrustLens AI
TrustLens AI will be enhanced with advanced features such as real-time URL and image analysis to improve threat detection. The platform will expand into a mobile app and browser extension for better accessibility. Future development will also focus on multi-language support, improved AI accuracy, and integration with educational institutions to promote cyber awareness at a larger scale.
Log in or sign up for Devpost to join the conversation.