(LLM) AI Generated Text Detection

Large Language Models have completely revolutionized text generation, with significant studies reporting an unprecedented quantity of human-quality content. Still, this super capability comes with some grave risks: the spreading of misinformation at an unprecedented scale, academic plagiarism, and even erosion of trust in written communication. We are therefore developing a robust AI-generated text detection system. Our two-phased approach first involves training a neural network classifier on at first carefully hand-crafted textual features designed to capture subtle variations between human and LLM-generated text, then creating a web application using Next.js that allows a user to easily input text to analyze and receive a clear classification outcome. This project is yet another contribution to combat the spread of misinformation and to the preservation of academic integrity. But most importantly, for the responsible and ethical use of powerful AI technologies. Keywords: AI-Generated Text, Text Detection, Large Language Models, Machine Learning, Academic Integrity, Misinformation, Authenticity