This Paper Was Written with the Help of ChatGPT: Exploring the Consequences of AI-driven Academic Writing on Scholarly Practices

Hongming Li, Seiyon Lee, Anthony F Botelho

Abstract: Recent advances in the development of large language models (LLMs) have led to power innovative suites of generative AI tools that are capable of not only simulating human-like-dialogue but also composing more complex artifacts, such as social media posts, essays, and even research articles. While this abstract has been written entirely by a human without any input, consultation, or revision from a generative language model, it would be difficult for a reader to discern the difference. Aside from some notable risks, questions remains as to how we should consider the originality of human work that are influenced or partially refined by a generative language model. We present this paper as both a case study into the usage of generative models to support the writing of academic papers but also as an example of how open science practices can help address several issues that have been raised in other contexts and communities. This paper neither attempts to promote nor contest the use of these language models in any writing task. The goal of our work to provide insight and guidance into the ethical and effective usage of these models within this domain.