AI Detector

A simple tool to detect whether a given text is AI-generated or human-written. This MVP uses a pre-trained model from Hugging Face's transformers library to classify text.

---

Features

• AI Detection: Detects if a given text is likely generated by an AI model (e.g., GPT).
• Confidence Score: Provides a confidence score for the prediction.
• Easy to Use: Simple Python function for quick integration.
• User Interface Enhancements:
  • Improved layout with a menu bar for easy navigation
  • Progress bar for batch processing
  • Copy results to clipboard
  • Export results to CSV
  • Clear functionality to reset interface
• Settings Dialog: Users can now configure security settings and UI preferences.
• Logging: Comprehensive logging of security events and analysis results.
• Batch Processing: Analyze multiple text files in a selected directory.

Security Features

• Input validation and sanitization.
• Rate limiting and file type restrictions.
• Comprehensive logging of security events and analysis results.
• Configurable limits for text length and file size.

---

Installation

Prerequisites

• Python 3.7 or higher
• pip (Python package manager)

Steps

1. Clone this repository:

   git clone https://github.com/se7enb2st/AI-Detector.git
   cd anti-ai-detection

2. Install the required Python libraries:

   pip install transformers torch
   

How It Works

• The tool uses the roberta-base-openai-detector model from Hugging Face, which is fine-tuned to detect text generated by OpenAI models like GPT. The model analyzes the input text and returns:
• Label: AI or Human
• Confidence Score: A value between 0 and 1 indicating the model's confidence in the prediction.

User Interface Features

1. Text Input Tab:

   • Enter text directly in the text area
   • Analyze text with a single click
   • Clear input and results
   • View results with confidence score

2. File Input Tab:

   • Select and analyze individual text files
   • Clear file selection and results
   • View detailed analysis results

3. Batch Processing Tab:

   • Process multiple text files in a directory
   • Progress bar shows analysis status
   • Copy results to clipboard
   • Export results to CSV format
   • Clear all results and reset interface

Example Output:

Prediction: AI (Confidence: 0.92)
Prediction: Human (Confidence: 0.85)