The Illusion of Self-Directed Learning? Rethinking the Impact of Generative AI on Critical-Spiritual Reasoning and the Tradition of Academic Mujahadah

Abstract

This study re-examines claims that Generative Artificial Intelligence (GenAI) necessarily strengthens self-directed learning in higher education. Rather than treating autonomy as independence in obtaining answers, the article defines genuine learning autonomy in terms of intellectual agency, verification, reflection, and responsibility for knowledge. Using a qualitative conceptual library-research design, the study brings recent international literature on GenAI, self-regulated learning, cognitive offloading, assessment, and AI ethics into dialogue with Islamic educational thought. The source corpus emphasizes peer-reviewed studies published between 2021 and 2026 while retaining earlier foundational works where conceptually necessary. Qualitative content analysis and comparative interpretation were used to identify recurring tensions concerning epistemic authority, cognitive delegation, desirable difficulties, and academic character. The synthesis suggests that GenAI can support learning when it functions as a scaffold for exploration, feedback, and metacognitive regulation, but excessive delegation of reasoning may create an appearance of autonomy without equivalent growth in intellectual agency. To address this tension, the article proposes an Academic Mujahadah Framework that positions GenAI as a wasilah (instrument), not a ghayah (end), and connects human verification, process-oriented assessment, and tabayyun with the preservation of adab and scholarly responsibility. The framework is conceptual rather than empirically validated; its value therefore lies in offering a testable ethical-pedagogical orientation for future higher-education research and practice.