Understanding the Practices, Perceptions, and (Dis)Trust of Generative AI among Instructors: A Mixed-methods Study in the U.S. Higher Education

TL;DR

This paper investigates how instructors in U.S. higher education perceive and engage with Generative AI (GenAI), focusing on practices, attitudes, and the dynamics of trust and distrust. Using a mixed-methods design with a survey of 178 instructors at a Mid-Atlantic university (March 2024) plus qualitative open-ended responses, the study finds a gap between familiarity with GenAI concepts and actual use in instructional tasks, and demonstrates that trust and distrust are related yet distinct constructs that can co-exist. The results show that familiarity tends to align with higher trust, while distrust often correlates with limited familiarity, with notable differences across teaching levels (undergraduate vs. graduate). The authors discuss implications for platform design, training, and policy to calibrate (dis)trust, highlighting ethical and environmental concerns, and advocating evidence-based interventions to integrate GenAI into teaching while maintaining academic integrity and learning outcomes.

Abstract

Generative AI (GenAI) has brought opportunities and challenges for higher education as it integrates into teaching and learning environments. As instructors navigate this new landscape, understanding their engagement with and attitudes toward GenAI is crucial. We surveyed 178 instructors from a single U.S. university to examine their current practices, perceptions, trust, and distrust of GenAI in higher education in March 2024. While most surveyed instructors reported moderate to high familiarity with GenAI-related concepts, their actual use of GenAI tools for direct instructional tasks remained limited. Our quantitative results show that trust and distrust in GenAI are related yet distinct; high trust does not necessarily imply low distrust, and vice versa. We also found significant differences in surveyed instructors' familiarity with GenAI across different trust and distrust groups. Our qualitative results show nuanced manifestations of trust and distrust among surveyed instructors and various approaches to support calibrated trust in GenAI. We discuss practical implications focused on (dis)trust calibration among instructors.

Figures (7)

• Figure 1: Response distribution of current practices and future intentions regarding the use of GenAI tools for instruction, ordered by the percentage of respondents indicating 'No plans to incorporate,' in ascending order.
• Figure 2: Response distribution of the types of instructional tasks for which instructors typically use GenAI tools, ordered by the percentage of respondents indicating 'Never,' in ascending order. For statements with the same percentage of 'Never,' ties are broken by the percentage of 'Rarely,' also in ascending order.
• Figure 3: Instructors’ attitudes towards training in GenAI, ordered by the percentage of responses in the most frequent category for each statement.
• Figure 4: Correlation matrix between all items for the trust and distrust scales to show how various trust (e.g., competence, personalization, etc.) and distrust (e.g., malevolence, dishonesty, etc.) attributes were statistically related. Each cell shows the correlation coefficient between two attributes, with values closer to 1 indicating a strong positive correlation and values closer to -1 indicating a strong negative correlation. The color gradient from blue to white represents increasing positive correlations, with blue denoting higher positive values and white indicating no or negative correlation.
• Figure 5: Trust and distrust distribution. Each point on the plot represents an instructor's score for trust (X-axis) and distrust (Y-axis), on a scale from 1 to 5.