Perceived Utility and Knowledge As Predictors of Generative AI Use Among Emerging Social Workers

Background:

As generative Artificial Intelligence (AI) continues to expand across academic and professional fields, its integration into ethical, human-centered professions like Social Work remains complex. While STEM disciplines have rapidly adopted AI tools, limited research explores how emerging Social Workers engage with these technologies in the United States. This study addresses that gap by examining how perceived utility and knowledge of AI influence usage patterns among emerging Social Work practitioners. Given Social Work’s ethical framework and relational practice model, understanding these predictors is essential for responsible and informed adoption of AI in both education and practice.

Methods:

A cross-sectional survey was administered to a sample of 97 emerging Social Workers at a southeastern United States university. The survey included 15 Likert-scale items measuring attitudes toward AI, alongside self-reported measures of perceived utility, knowledge, and frequency of use. The survey also included demographic information regarding the participants age. A two-step analytic strategy was used. First, a Principal Components Analysis was conducted to assess the factor structure of the attitude items and ensure dimensional validity. Second, a multiple regression model was performed with frequency of AI use as the dependent variable and perceived utility, knowledge, and age as independent variables. An interaction term between utility and knowledge was also included to determine if moderation was present between the knowledge and utility constructs. Age was used as a control variable within the regression.

Results:

The Principal Component Analysis supported the integrity of the Likert-scale items, identifying clear dimensions of AI attitudes. Regression modelling showed that both perceived utility (β = 1.88, p <.001) and AI knowledge (β = 1.63, p <.05) were significant predictors of use, indicating that emerging Social Workers engage with AI tools more frequently when they find them useful and feel knowledgeable about them.

Implications:

These findings suggest that perceived utility and self-assessed knowledge are key drivers of AI adoption among emerging Social Work practitioners. Contrary to expectations, prior knowledge of AI did not enhance the effect of perceived utility. This implies that students with limited experience may still adopt AI tools if they perceive them as beneficial. These insights underscore the necessity of a training design that highlights the practical utility of AI while simultaneously imparting knowledge on how to effectively utilize it as a tool. By fostering responsible engagement with emerging technologies, those responsible for training Social Workers can help to prepare future practitioners to navigate an evolving digital landscape while upholding the core values of the profession.