Integrating Generative AI within Deep Learning Frameworks to Transform Student Learning Dimensions: A Systematic Literature Review

Abstract

The rapid advancement of Artificial Intelligence (AI) in education has increased the necessity of understanding its role in supporting learning processes. Most current studies focus on teacher performance or instructional efficiency, while the relationship between AI and the student learning model remains under-explored. This study aims to analyze how the utilization of AI supports student learning models and identifies specific AI technologies, applications, or platforms that optimize these models. Through a Systematic Literature Review (SLR) protocol tracking 392 initial records down to 25 final articles from Scopus and SINTA (2020-2025) databases, this study provides an explicit methodological framework mapping AI's interaction with student agency. Theoretically, it shifts the educational discourse from teacher performance to student-centered autonomy, while practically offering design strategies for implementing collaborative AI tools within deep learning environments. Through a SLR method, two primary findings emerged. First, AI technologies such as Intelligent Tutoring Systems (ITS) and Large Language Models (LLM) significantly influence indicators of aspirations, learning to learn capabilities, and interpersonal relationships. Second, AI-based learning environments facilitate a more adaptive, personalized, and student-centered learning model by simultaneously integrating cognitive, social, and motivational dimensions. These findings confirm that AI is not merely an automation tool but a collaborative partner that strengthens student learning autonomy.