Write incentives? Authors’ sentiment on generative AI
Year: 2026
Type: article
Study design: Observational Study
Abstract: Generative AI is reshaping the book industry, raising critical questions about intellectual property and fair compensation. While legal and general user perspectives have been widely examined, authors’ views remain under-explored. This study addresses that gap by examining whether authors are willing to allow their works to be used in AI training and what compensation they expect for existing and future works. Using survey data from members of the Australian Society of Authors, we apply Seemingly Unrelated Bivariate Probit and Seemingly Unrelated Regression models to identify key factors shaping these decisions. Results show that authors’ willingness to allow AI training is influenced by their fairness perceptions, prior beliefs about AI training, and patterns of AI use, while compensation expectations are primarily driven by concerns about replacement risk and income loss. These findings contribute new evidence to debates on AI, copyright, and creative labour markets.
Source: Applied Economics
Authors: Shujie Liang, Paul Crosby, Jordi McKenzie
Institution: Macquarie University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7165169031