Open Access

Melanie Raphaela Stöckl

• The increasing use of Artificial Intelligence (AI) in higher education is transforming students’ learning practices by shifting planning, monitoring, and evaluation toward AI-supported interaction. While task performance may remain high, externalizing regulatory processes may limit the development of self-regulated learning (SRL). This study examines how students appropriate dialogic AI support within their SRL practices while preserving regulatory agency. Drawing on a sociotechnical perspective, AI appropriation is conceptualized as meta-regulation within the SRL cycle. Survey data and think-aloud-informed group discussions were analyzed to explore associations between baseline SRL profiles, reported changes, and situated AI use. Correlation analyses indicate exploratory, profile-contingent patterns, with lower baseline SRL tending to coincide with larger reported gains. Qualitative findings illustrate how students enact AI as a co-regulatory resource for SRL and critical boundaryThe increasing use of Artificial Intelligence (AI) in higher education is transforming students’ learning practices by shifting planning, monitoring, and evaluation toward AI-supported interaction. While task performance may remain high, externalizing regulatory processes may limit the development of self-regulated learning (SRL). This study examines how students appropriate dialogic AI support within their SRL practices while preserving regulatory agency. Drawing on a sociotechnical perspective, AI appropriation is conceptualized as meta-regulation within the SRL cycle. Survey data and think-aloud-informed group discussions were analyzed to explore associations between baseline SRL profiles, reported changes, and situated AI use. Correlation analyses indicate exploratory, profile-contingent patterns, with lower baseline SRL tending to coincide with larger reported gains. Qualitative findings illustrate how students enact AI as a co-regulatory resource for SRL and critical boundary setting. These findings suggest that AI’s educational value may depend on how learners configure AI within their regulatory routines.…