Primary Research Group Releases New Report: Survey of Library Science Faculty: Contributions of Content to AI Models, ISBN 979-8-88517-323-0
Primary Research Group has published a new report, Survey of Library Science Faculty: Contributions of Content to AI Models, the first systematic look at how library science faculty are contributing—voluntarily or otherwise—to artificial intelligence training data, and how they perceive the use, reuse, and risks associated with their scholarly and instructional materials.
The report provides extensive data tables, faculty subgroup analysis, and open-ended commentary that together map a rapidly evolving relationship between academic content creation and AI model development. Findings cover issues ranging from voluntary submissions to concerns about unauthorized use, departmental activities, and early classroom experimentation with AI-assisted tools.
Some Key Findings
More than half of faculty believe they may have content suitable for training AI models — or are unsure.
55.55% of respondents either said they have AI-trainable content (24.44% “Yes”) or are unsure (31.11% “Not really sure”).
Just 6.67% of faculty reported submitting articles, data, or instructional materials for use in AI training (Table 2.1).
Nearly 29% say “Yes” when asked whether their content has been used in AI models without permission, and uncertainty is widespread
11.11% of faculty report using their own class materials—notes, videos, research, or texts—in a teaching chatbot or model.
Just 8.89% say their department has taken steps to collect or prepare faculty content for AI use, and several respondents describe opaque or revenue-driven departmental motives.
About the Report
Survey of Library Science Faculty: Contributions of Content to AI Models includes:
-
150+ tables (sample dependent) breaking down responses by:
-
Institutional rank
-
Data Broken Out by Carnegie Classification
-
Data Broken Out by Enrollment size
-
Data Broken Out by Faculty Rank
-
Data Broken Out by Age, Gender, and Political Views
-
Data Broken Out by Sector (public vs. private)
-
Data Broken Out by Level of Teaching Load
-
Verbatim open-ended responses illustrating concerns about unauthorized use, shifting expectations, and emerging instructional experimentation
-
Analysis of faculty uncertainty regarding obligations, rights, and opportunities as AI developers and universities seek new sources of training data
This report is essential reading for library schools, academic departments, faculty leaders, publishers, university administrators, and anyone seeking to understand how the emergence of AI models intersects with academic authorship and scholarly rights.
Availability