Summary
The claim made in this open access paper is quite remarkable and important, if accurate. We haven’t tried the method ourselves yet, but we can see how it will be of value to teachers, tutors and editors. One wonders how long it will be before the system designers come up with a way to beat the detection method.
Desaire, H., Chua, A.E., Isom, M. Jarosova, R. & Hua, D. (2023) Distinguishing academic science writing from humans or ChatGPT with over 99% accuracy using off-the-shelf machine learning tools. Cell Reports Physical Science, 4(6)
Publisher (Open Access): https://www.sciencedirect.com/science/article/pii/S266638642300200X?via%3Dihub
Publisher (Open Access): https://www.sciencedirect.com/science/article/pii/S266638642300200X?via%3Dihub