The rapid adoption of generative artificial intelligence (AI) in higher education has intensified a pedagogical dilemma: while AI tools can increase immediate classroom engagement, they do not necessarily foster the self-regulated learning (SRL) capacities required for ethical and reflective professional practice, particularly in human-service fields. In this two-time-point, pre-post cohort-level (repeated cross-sectional) evaluation, we examined a six-week AI-integrated curriculum incorporating explicit SRL scaffolding among social work undergraduates at a Taiwanese university (pre-test N = 37; post-test N = 35). Because the surveys were administered anonymously and individual responses could not be linked across time, pre-post comparisons were conducted at the cohort level using independent samples. The participating students completed the AI-Enhanced Learning Attitude Scale (AILAS); this is a 30-item instrument grounded in the Technology Acceptance Model, Attitude Theory and SRL frameworks, assessing six dimensions of AI-related learning attitudes. Prior pilot evidence suggested an engagement regulation gap, characterized by relatively strong learning process engagement but weaker learning planning and learning habits. Accordingly, the curriculum incorporated weekly goal-setting activities, structured reflection tasks, peer accountability mechanisms, explicit instructor modeling of SRL strategies and simple progress tracking tools. The conducted psychometric analyses demonstrated excellent internal consistency for the total scale at the post-test stage (Cronbach’s α = 0.95). The independent-samples t-tests indicated that, at the post-test stage, the cohorts reported higher mean scores across most dimensions, with the largest cohort-level differences in Learning Habits (Cohen’s d = 0.75, p = 0.003) and Learning Process (Cohen’s d = 0.79, p = 0.002). After Bonferroni adjustment, improvements in the Learning Desire, Learning Habits and Learning Process dimensions and the Overall Attitude scores remained statistically robust. In contrast, the Learning Planning dimension demonstrated only marginal improvement (d = 0.46, p = 0.064), suggesting that higher-order planning skills may require longer or more sustained instructional support. No statistically significant gender differences were identified at the post-test stage. Taken together, the findings presented in this study offer preliminary, design-consistent evidence that SRL-oriented pedagogical scaffolding, rather than AI technology itself, may help narrow the engagement regulation gap, while the consolidation of autonomous planning capacities remains an ongoing instructional challenge.

While the use of artificial intelligence (AI) in higher education is widespread, students’ experiences with AI-supported learning in their regular courses remain underexplored. Objective: This research examines the relationships among perceived usefulness, perceived ease of use, and academic engagement among undergraduate students enrolled in AI-supported courses at a Taiwan university. It adopts the Technology Acceptance Model, where learning desire indicates perceived usefulness, and technology self-efficacy indicates perceived ease of use. Methods: The study takes a questionnaire with six dimensions of technology self-efficacy, learning desire, learning methods, learning planning, learning habits, and learning process to evaluate students’ attitudes toward AI-supported learning and their academic engagement. Results: Students’ attitudes toward AI-supported learning were moderate to positive. Multiple regression analysis showed that perceived usefulness was significantly and positively associated with academic engagement, whereas perceived ease of use showed a positive but non-significant association. Implications: Students’ academic engagement is influenced more by how useful AI tools are perceived for learning, rather than by their confidence in using AI tools. This paper enriches the literature on student-centered AI in higher education and gives insights for designing AI-supported courses that integrate AI tools with meaningful learning tasks. Future research can examine larger and more diverse samples and use longitudinal or experimental designs to test how students’ perceptions of AI tools develop over time.

Artificial intelligence (AI) has become increasingly embedded in technology-enhanced learning environments, where peer assessment now serves both instructional and analytic purposes. Beyond allocating feedback and grades, it also produces data that is later interpreted through learning analytics systems. In practice, visible indicators such as students’ fairness perceptions and the degree of agreement among peer raters are often treated as signs that the assessment process is functioning effectively. However, these indicators do not necessarily correspond to grading validity. Students may regard a peer assessment process as fair even when peer-generated ratings remain weakly aligned with expert judgement. This study, therefore, examines whether the socio-technical configurations associated with high perceived fairness in a digitally mediated peer assessment environment also correspond to criterion-referenced grading validity. Data were collected from 215 undergraduate students enrolled in an Artificial Intelligence Foundations course over two consecutive semesters at a university in Taiwan, with instructor ratings serving as an external expert reference within the course context, rather than as a universal ground truth. Because anonymity conditions and semester were fully confounded in the study design, differences linked to anonymity should not be interpreted as isolated causal effects. A two-stage fuzzy-set Qualitative Comparative Analysis (fsQCA) was used. In the first stage, three equifinal configurations associated with high perceived fairness were identified. In the second stage, these configurations were examined against four grading objectivity outcomes: peer–instructor alignment, peer convergence, familiarity bias, and leniency bias. The findings show that fairness perception and grading validity are only partially aligned. Configurations anchored in explicit criterion transparency consistently supported both experiential legitimacy and evaluative accuracy. By contrast, one configuration was associated with high peer convergence while remaining weakly aligned with instructor standards, a pattern described here as false objectivity; this context-dependent configurational finding warrants further investigation across other settings. The study contributes to research on digitally enhanced assessment and learning analytics by showing that fairness perception, peer convergence, and grading validity should be treated as analytically distinct dimensions of assessment quality.

This exploratory research examines learning-process engagement in artificial intelligence (AI)-supported courses, measured through students’ self-reported attention, active participation, and thinking during class. Necessary Condition Analysis and fuzzy-set qualitative comparative analysis were applied to questionnaire data from three cohorts at a Taiwanese university of technology, using 90 cases after screening 107 responses for invariant responding. Seventeen respondents selected the same option across all 30 items and were excluded from the main analysis. The study examines five measured learner characteristics and study practices: academic self-concept, learning desire, learning strategies, learning planning, and habitual regulation. In the screened sample, no condition meets the set-theoretic necessity criterion, while four show statistically significant NCA ceiling effects. High engagement is associated with one configuration combining learning desire, strategies, planning, and habitual regulation, with strategies and habitual regulation as core conditions. Low engagement is associated with the joint absence of all five conditions. Among all 107 respondents, no condition met the set-theoretic necessity criterion. Re-estimating the calibration anchors yielded five high-engagement configurations, while holding the original anchors fixed retained the single configuration. For instructors, these patterns offer questions for future study rather than validated rules for choosing support for individual students.