Needs assessment and technical specification for an ai feedback system for studio-based learning

Studio- based learning (SBL) learning environments, typical of design and creative disciplines, rely on formative feedback and critique. Yet, it is difficult for educators to provide timely, substantive feedback to each student, and students report delay and brevity in comments. The current study fills that void by conducting needs assessment and conceptualizing an AI-enabled feedback system to complement studio-based learning. The study employed mixed-methods design: educator focus group interviewing (N=11) in Kenya, Saudi Arabia, and USA, and student survey (N=150) of studio classes. Findings indicate widespread frustrations about turn-around time to provide comments, comments depth, and grading workload. Kenyan, Saudi Arabian, and American instructors described feeling bogged down by high class enrollments and grading burden, as is also seen in studies on rising student-staff ratios and comments lag time. Students also described waiting weeks to receive feedback and that the comments largely lacked depth to enact change, as also seen in previous studies on discontent with feedback. Each of these groups also expressed optimism that an AI-enabled feedback system could provide more timely, individualized critiques, provided it is well-designed. The principal deliverable of this study is an elaborated technical specification sheet (see Appendix) of features and structure of conceptualized AI system. The study describes functional requirements (e.g., an AI engine to generate rubric-matched formative comments), technical design (integration into learning platforms, data safeguarding), and collaboration plans to engage a software developer and development cost estimate. Finally, the study proposes an orderly process of future development: sequential pilot testing, user feedback-led fine-tuning, and rollout to scale up. This project, financed by a $10,000 Stephen F. Austin State University JACK STARS grant, paves the ground for closing the pedagogical disparity of AI in SBL feedback.