Beyond Faster Learning: How GenAI Learning Efficiency Translates into Effective Learning Outcomes

Generative artificial intelligence (GenAI) can make learning faster and less demanding, but greater efficiency does not necessarily result in better learning. This study examines how GenAI learning efficiency is translated into effective learning outcomes among higher-education students in India. GenAI Learning Efficiency (GLE) is conceptualized as a higher-order construct comprising learning time efficiency, cognitive effort efficiency, learning task efficiency, and knowledge access efficiency. Drawing on Cognitive Load Theory and Self-Regulated Learning theory, the study examines cognitive engagement and self-regulated learning as parallel mechanisms connecting GLE with effective learning outcomes. Data were collected from 548 undergraduate and postgraduate students with prior experience using GenAI for academic purposes and analysed using PLS-SEM. The findings show that GLE strengthens both cognitive engagement and self-regulated learning, which subsequently contribute to effective learning outcomes. Both mechanisms mediate the relationship between GLE and learning outcomes, while a smaller direct effect remains. The study shows that the educational value of GenAI lies not simply in making learning easier or faster, but in whether learners use the resulting efficiency to engage more deeply and regulate their learning effectively.