Integration of a Personalized Generative AI into University Instructional Design: Contributions, Uses, and Conditions for the Effectiveness of Active Learning Methods

The rapid expansion of generative artificial intelligence (AI), particularly large language models (LLMs), is profoundly transforming higher education by enabling the on-demand production of instructional content, learning activities, feedback, and assessment tools. However, recent research indicates that these uses remain largely opportunistic and insufficiently embedded within systematic instructional design processes. This weak integration may compromise constructive alignment between intended learning outcomes, learning activities, and assessment methods, as well as undermine academic integrity. And this article offers a structured synthesis of the contributions, uses, and effectiveness conditions of a personalized generative AI—defined as an AI system configured through stable pedagogical instructions, institutional constraints, and disciplinary frameworks—designed to support active learning approaches in higher education. The study is based on a qualitative documentary analysis of recent international scientific publications (during 2023–2025), complemented by the examination of a university teaching resource used as an empirical case of implementation. The findings highlight three major contributions. First, personalized generative AI can support the entire instructional design cycle, from needs analysis to assessment design, by strengthening pedagogical coherence and constructive alignment. Second, the dominant uses identified primarily concern the design of active learning strategies (flipped classroom, case-based learning, structured debates, collaborative projects) and the enhancement of formative and summative assessment practices, in line with empirical evidence demonstrating the positive impact of active learning on student performance and success. Third, the effectiveness of these uses depends on key conditions: the development of AI literacy and prompt engineering skills among educators and students; the redesign of assessment systems to ensure robustness against automation; and the establishment of ethical and institutional governance grounded in recognized risk management frameworks and international guidelines for AI in education.