Publications in the Age of Generative AI

Music City Center 

Room: CC-104C 

Generative AI has exploded in popularity and notoriety since November 2022. At JSM 2023 there were only 2 sessions on generative AI, and at JSM 2024 there were close to 30. Statisticians and data scientists are right to be involved in the ongoing discussion of the appropriate use of generative AI, particularly when it comes to academic writing. In May of 2023, Wiley discontinued 4 scientific journals due to association with paper mills, many of whom use generative AI to write entire papers. ArXiv preprint service reported that the volume of papers in several categories increased dramatically between 2019 and 2023 – a period that roughly coincides with the debut of tools like ChatGPT. As academic papers are the currency of science and the basis for promotion and tenure, the issue of the appropriate use of generative AI in academic writing is important. The theme of JSM 2025 is "Statistics, Data Science, and AI Enriching Society". The invited panelists in this session will give their perspectives on how generative AI can enrich academic publication and the statistics profession. We have gathered a panel of experts to talk about the salient issues of the use and misuse of generative AI in publication, and the panel will address both AI policies and measures in place to enforce them.

John Aston is currently Harding Professor for Statistics in Public Life, and Pro-Vice-Chancellor for Research at the University of Cambridge. He will be President of the Royal Statistical Society in 2025-26. He has worked academic and government positions, and has been an associate editor for quite a few journals both in statistics and other professions. He sees generative AI as a tool, something that we will need to work with, and it will become increasingly important for statisticians to be part of the conversation about what can be our role in helping determine the use of AI in publications, not just in statistics but in wider fields too.

Paul Naish will offer his perspective as the Global Head for Portfolio for Taylor & Francis journals covering Mathematics, Statistics, Data Science, History of Science, and Science, Technology & Society. Taylor & Francis is the publisher for ASA journals. He has worked closely with the ASA and other societies in developing policies for the use of generative AI in scientific publications.

Dan Nettleton will offer his perspective as a department chair advising faculty and graduate students on the use of generative AI in the writing of dissertations and articles. He is Distinguished Professor, Laurence H. Baker Endowed Chair in Biological Statistics, and Chair of the Department of Statistics at Iowa State University. Nettleton is past chair of the American Statistical Association (ASA) Caucus of Academic Representatives. From 2020 to 2024, Nettleton served as secretary of the Statistics Section (Section U) of the American Association for the Advancement of Science (AAAS) and now is Section U's representative on the National Institute of Statistical Sciences Board.

Juana Sanchez is Senior Lecturer Emerita at UCLA Department of Statistics and Data Science. She will add her expertise as Editor of the Journal of Statistics and Data Science Education and educator. She will also address the positive aspects of AI, such as the ability to improve workflow and save time on mundane tasks, which are features of AI that teachers, learners, researchers, editors, authors and peer reviewers all value. However, there are limits to how these features can be used by academic journal authors without breaching journal policy.

Yes

Committee on Publications

ASA Caucus of Academic Representatives