An all-in-one web platform for real-time fact-checking, deepfake detection, and tracking the latest misinformation trends.
inCredible AI is a powerful web platform designed to combat the spread of online misinformation. It integrates real-time fact-checking, deepfake detection, and a curated display of the top three recent misinformation trends—all in one user-friendly interface.
- Frontend: React
- Backend: Flask
- Database: AWS S3
- AI/ML: OpenAI API, Serpapi API, Gemini API, Arya AI API, Google Fact Check API, YOLO model, DALL E 3 API, AWS Textract
- Hosting: AWS EC2
- 🔍 Real-time Fact-Checking: Verify claims instantly using cutting-edge AI APIs.
- 🕵️ Deepfake Detection: Identify manipulated media with the YOLO model.
- 📈 Misinformation Tracker: Displays the top 3 trending misinformation stories.
- 🌐 Web Scraping: Pulls data dynamically with BeautifulSoup4 for up-to-date insights.
- Node.js & npm installed
- Python
- Required API KEYS:(SERP API, Google Fact Check API, OpenAI API, GEMINI API, ARYA API, AWS ACCESS KEY, AWS SECRET ACCESS KEY, AWS Region)
- Clone the repo
git clone https://github.com/fearyj/inCredible-AI.git - Create python virtual environment
python -m venv venv - Activate virtual environment
For Windows:
venv/Scripts/activate
For Mac:
source venv/bin/activate
-
Change directory to frontend
cd frontend -
Install npm libraries
npm install -
Edit the .env file to include your Keys
-
Run frontend
npm run dev -
Run backend
cd ../backend
python App.py -
NOTES: Change origins port number to the one shown in terminal when running frontend
Meet the amazing team behind this project! 🎉
| Name | Role | GitHub / LinkedIn |
|---|---|---|
| Tan Yoong Ken | Backend Developer | GitHub / LinkedIn |
| Hoo Jing Huan | Backend Developer | GitHub / LinkedIn |
| Tan Peng Yee | Frontend Developer | GitHub / LinkedIn |
| Rachel Phuar Yi Ling | Frontend Developer | GitHub / LinkedIn |
We have exciting plans to improve this project! Here are some features we'd love to implement:
- 🔹 Multi-language support - Expand fact-checking and deepfake detection to multiple languages.
- 🔹 Blockchain verification - Store fact-checked results on a blockchain for greater transparency.
- 🔹 User-submitted fact-checks - Allow users to submit and vote on misinformation claims.
- 🔹 Mobile App Version - Develop a mobile version of the tool for easier access.
- 🔹 Deepfake Detection from URL - Allow users to paste in Youtube URL to check for deepfake.
For questions, reach out to us at:
- Ken - ytan244@e.ntu.edu.sg
- Jing Huan - hooj0006@e.ntu.edu.sg
- Peng Yee - tanp0124@e.ntu.edu.sg
- Rachel - RPHUAR001@e.ntu.edu.sg
✨ Made with ❤️ at [TechFest - 2025] ✨