Author: Almira Osmanovic Thunström
Publisher: Scientific American
Publication Year: 2022
Summary: The following article describes how GPT-3 is, according to Wikipedia, “…an autoregressive language model that uses deep learning to produce human-like text.” The author of this article gave the model a short but complex query: “Write an academic thesis in 500 words about GPT-3 and add scientific references and citations inside the text.” Within two hours the task was completed. The ethical considerations of the algorithm having done this began to crop up when Ms. Thunström began the process of submitting the paper for academic peer review. Questions of whether or not an AI, however sophisticated it may be, is still a tool, whether consent is needed or not before the work is published, and if it as the author has any conflicts of interest were proposed. Following these questions, additional concerns about repercussions should the paper actually be published were proposed, such as must a person now prove that any writing that they submit for publication was not created with significant input from an algorithm?