2026 ASEE Annual Conference & Exposition

Student success remains a central focus in higher education, with institutions, faculty, students, and policymakers defining it through different lenses, from quantitative metrics like retention and graduation rates to qualitative dimensions of learning, belonging, and transformation. As generative artificial intelligence (AI) systems become embedded in academic environments, their role in shaping how success is understood and pursued has become increasingly consequential. Students, faculty, and administrators now use AI tools to assist with completing assignments, designing curricula, providing feedback, drafting institutional policies, and generating analytical reports, tasks that directly influence pedagogical practice and institutional decision-making. These systems synthesize vast corpora of academic literature, policy documents, institutional discourse, and popular media, potentially amplifying dominant narratives about educational success while marginalizing alternative perspectives. Yet little research examines what conceptualizations of student success these AI systems encode, whose values and assumptions they privilege, or how their framings might reshape educational practice if adopted uncritically.

This study employs grounded theory methodology to analyze conversational data generated by multiple large language models prompted to adopt the role of an educational philosopher while responding to structured interview questions about student success in higher education. Open and axial coding will be used to identify recurring themes, assumptions, and value frameworks embedded within AI-generated narratives. The goal of the study is to construct a theoretical model that explains how generative AI represents student success and to explore the potential implications of these representations for advising, assessment, and curricular decision-making. The analysis will also examine whether AI-based conceptions of success align with or diverge from human-centered perspectives common in engineering education literature. By examining how AI systems interpret and articulate the meaning of success, this study aims to illuminate the cultural and epistemological assumptions that may become embedded in future educational technologies. The anticipated outcome is a deeper understanding of how generative technologies might influence institutional narratives of success and how these definitions, if adopted uncritically, could shape the values, goals, and practices that lie beneath the teaching and learning of engineering in higher education.