This article examines internationally mobile students' engagement with the ubiquity of AI technologies in U.S. higher education. Drawing on the Emirbayer and Mische (1998) ecological approach to agency, the study asks how multilingual graduate students' iterational, practical-evaluative, and projective dimensions of agency emerge to shape their academic practices and decision-making regarding AI use. Data were derived from an online screening questionnaire, individual interviews, and technology-related artifacts from 14 students at a U.S. university. Analysis of the interview data, with participant-provided screenshots as elicitation materials during the interviews, reveals how their recollections, contextual evaluations, and future goals were manifested in their reported AI practices. The findings suggest that multilingual international graduate students' AI use was not oriented exclusively toward the past, present, or future; rather, their agency emerged through the interpenetration of all three temporal dimensions. Participants' decisions about whether, when, and how to use AI reflected situated judgments in which prior experiences and established practices informed responses to immediate academic demands while anticipated futures simultaneously shaped the boundaries they placed around AI use. Rather than simply adopting or rejecting AI, participants further demonstrated agentic dilemmas in AI use by negotiating between the efficiency of AI versus deprived opportunities for deep learning and independent thinking, using AI for support versus withholding AI to avoid overreliance, and using one's own voice versus conforming to standard English norms. The findings highlight the temporal-relational nature of these students' agentic judgments and strategies in relation to their graduate studies in U.S. higher education.
No takes yet. Share an insight, caveat, or question.
Hyun‐Sook Kang (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: