Generative AI in research evaluation

About this video

Virtual panel co-sponsored by Cell Press and Elsevier recognizing Peer Review Week.

Generative AI tools like ChatGPT can create anything from poems to Python code. Researchers are increasingly using AI both as part of their research and to communicate and publish their work. These tools are powerful, but there are concerns with confidentiality and ethics, their output is not always accurate, and their operation can be difficult to understand. Join panelists Giovanni Cacciamani (University of Southern California), Andrew Hufton (Cell Press), Bahar Mehmani (Elsevier), and Lisa Rasmussen (University of North Carolina, Charlotte) as we discuss what actually happens when a researcher uploads content to ChatGPT, how to peer review research that incorporates generative AI in its methods, and whether there are any appropriate applications of AI in the peer review process.

About the presenters

Andrew Hufton
Andrew Hufton
|

Editor-in-Chief of Patterns, Cell Press

Andrew is the Editor-in-Chief of Patterns, a data-science-focused journal in the Cell Press family that publishes a wide range of machine learning and AI research. He has a passion for promoting open science and improving how we publish data-rich research. He has worked across multiple publishers and journals, including launching the data journal Scientific Data. He received his PhD from Stanford University in 2006 and, prior to his editorial career, conducted research on topics in developmental biology and genome evolution.

Bahar Mehmani
Bahar Mehmani, PhD
|

Peer Review Innovation Lead, Elsevier

Bahar is peer review innovation lead at Elsevier and oversees the peer review strategy and innovation for Elsevier journals. She is the chair of the recently launched Peer Review Workbench, a unique dataset allowing evidence-based research in peer review. She received her PhD in Theoretical Physics from the University of Amsterdam and moved to the Max Planck Institute for the Science of Light to pursue her academic career as a postdoc. She joined Elsevier in 2013 as a managing editor and since then has studied peer review as a scientific research topic. Bahar is vice president of the European Association of Science Editors, and co-chair of NISO's Inclusion, Equity, Diversity, and Accessibility committee.

Lisa Rasmussen
Lisa Rasmussen
|

University of North Carolina, Charlotte

Lisa earned her PhD in Philosophy with a focus in bioethics from Rice University in 2003. She is a Professor in the Department of Philosophy at UNC Charlotte, and currently serves as a Faculty Fellow in the Graduate School with a focus on fostering a campuswide culture of research integrity. She is the Editor-in-Chief of Accountability in Research. One of Dr. Rasmussen’s main research areas is research ethics, particularly research misconduct and federal regulatory approaches to ethical issues in research. She is presently working on a book manuscript on unregulated human subject research, motivated by her recent work in citizen science and DIY Bio. She also writes and teaches in the areas of healthcare ethics and ethics consultation. Dr. Rasmussen has been PI or Co-PI on over $1 million in NSF awards, and serves as and Co-Editor of the book series Philosophy and Medicine (Springer) comprising nearly 150 volumes.

Giovanni Cacciamani
Giovanni Cacciamani
|

University of Southern California

Giovanni is an Italian urological surgeon and physician-scientist. He currently serves as an Associate Professor of Urology and Radiology at the Department of Urology of the University of Southern California. He is also the Director of the Artificial Intelligence Center for Surgical and Clinical Application in Urology and the Vice-chair of the Research Council at the Department of Urology of USC. Dr. Cacciamani's primary areas of interest include Urological Oncological diseases such as Prostate Cancer, Renal Cancer, and Bladder Cancer. He is also deeply interested in patient safety and the application of Artificial Intelligence in healthcare, particularly in the use of machine learning for identifying radiomics features to predict pathology from imaging and in detecting predictors of recurrence in urological malignancies. Dr. Cacciamani is the Principal Investigator of several guidelines aimed at improving the reporting of AI intervention in healthcare, including PRISMA-AI. Most recently, he established on the CANGARU (ChatGPT, Generative Artificial Intelligence, and Natural Large Language Models for Accountable Reporting and Use) Guidelines. This initiative, developed in collaboration with editors from top-ranked academic journals, regulatory agencies, publisher representatives, and Artificial Intelligence experts, aims to establish guidelines for the ethical use of Generative AI (GAI), GPTs, and Large Language Models (LLMs) in academia. To date, Dr. Cacciamani has published 280 articles in international peer-reviewed journals and is the author of more than 200 abstracts presented at national and international meetings. He is part of the steering committee of several reporting guidelines aiming to enhance the quality of reporting in scientific and medical literature.

Matthew Pavlovich
Matthew Pavlovich (moderator)
|

Editor-in-Chief of Trends in Biotechnology, Cell Press

Matthew earned his BS in chemical engineering from the Georgia Institute of Technology and his PhD in chemical engineering from the University of California at Berkeley, where he studied the biological effects of air plasmas. He studied analytical chemistry as a postdoctoral researcher at Northeastern University, then joined Cell Press at the start of 2016. Matt is a senior manager in the Trends group and has also worked in secondments as the product manager for Cell Press Community Review and as the acting editor-in-chief of STAR Protocols. 

Comments