Reality Check is a web-based tool that helps users verify whether digital content (text or images) is likely to be real or AI-generated. The goal of the project is to promote digital awareness and support safer online behavior.
- Image authenticity detection
- Text authenticity analysis
- Simple and user-friendly web interface
- Clear explanations with confidence scores
The user interacts with a simple web interface where they can upload an image or paste text.
The request is sent to a backend server built using FastAPI.
The backend:
- Analyzes images using a pre-trained model
- Analyzes text using language-based heuristics
The results are converted into human-readable feedback.
- HTML
- CSS
- JavaScript
- Python
- FastAPI
- Uvicorn
cd reality_check_backend
pip install -r requirements.txt
python -m uvicorn main:app --reload
##
###
### Frontend
Open reality_check_frontend/index.html in any web browser.
Why This Project Is Important
Reality Check helps people become more aware of the content they consume online. Instead of blindly trusting information, users are encouraged to verify and think critically.
This can help in:
Reducing misinformation
Improving digital awareness
Promoting responsible sharing
Supporting safer online behavior
The project aligns with the idea of cyber peace by focusing on trust and transparency.
Limitations
The system provides probability, not certainty.
It may sometimes classify real content as risky or miss AI-generated content.
Accuracy depends on the quality of the models and input data.
Future Improvements
Support for video and audio deepfake detection
Improved model accuracy with larger datasets
User history and reporting dashboard
Browser extension integration
## Screenshots
