Co-Designing AI-Enabled Learning in Nursing Education: A Learning Engineering Approach Using Faculty and Student Insights

Generative AI tools are rapidly entering nursing education. Yet little empirical work examines how faculty and students actually use these tools in authentic learning contexts or what design requirements emerge from their lived experiences. This practitioner-oriented study investigates an academic-industry partnership through the adoption of Sherpath AI (SPAI), a generative AI chatbot for nursing education, across 45 prelicensure course sections at Arizona State University's Edson College of Nursing and Health Innovation. Through semi-structured interviews with seven faculty members and focus groups with ten students, we identified patterns in AI appropriation for studying, clinical reasoning, and pedagogical design. Drawing on learning engineering (LE) frameworks emphasizing human-centered, iterative, and contextually grounded design, we surface key themes: AI as a cognitive and pedagogical partner, trust calibration through evidence pathways, prompting as a metacognitive scaffold, and academic-industry learning loops. Findings reveal opportunities - personalized explanations, confidence building, and instructional coherence - and priorities, including verification pathways in specialized domains and scaffolded prompting support. We conclude with empirically derived design requirements for scalable, equitable, and trustworthy AI-enabled learning systems in safety-critical professional education.