Handwritten documents remain an essential medium for communication in academic, administrative, and social contexts. However, converting them into digital formats with high accuracy continues to present challenges, especially in cases where legibility is poor or writing styles vary significantly. Optical Character Recognition (OCR) technologies have been widely adopted to automate text extraction from printed and handwritten sources, but they often produce error-prone outputs when handling cursive or irregular handwriting. At the same time, the emergence of advanced generative Artificial Intelligence (AI) models, such as OpenAI’s GPT family, has enabled automated correction of textual errors while preserving meaning. This article presents OCRFlow, a novel hybrid system that combines Google Cloud Vision OCR and OpenAI’s GPT with a user-friendly graphical interface built using Python’s Tkinter. The study follows a combination of Object Oriented Analysis and Development (OOAD) and Design Science Research (DSR) Methodologies, involving iterative prototyping and user feedback to improve usability and accuracy. A series of evaluations were conducted to assess OCR accuracy, correction performance, usability, and efficiency. OCRFlow provides a practical proof of concept for hybrid AI-OCR systems that enhance digital transformation of handwritten content. The article contributes to both research and practice by demonstrating how existing AI services can be orchestrated into a cohesive tool that addresses real-world problems in document management. It also highlights the broader implications of integrating OCR with LLM-based correction for educational institutions, government offices, and industries where handwritten records remain prevalent.

Keywords: Artificial Intelligence, OCR, GPT, deep learning, machine learning, language model, transformer model