Abstract
<title>Abstract</title> <p>Background Artificial intelligence (AI) is increasingly transforming mental health care through emotion recognition, digital phenotyping, wearable technologies, chatbots, and decision-support applications. However, many AI education programs remain directed toward technically trained learners, while health science and non-STEM students have fewer opportunities to develop applied AI competencies. This study evaluated the short-term outcomes of an interdisciplinary AI and mental health prototype development program for health science and non-STEM students. Methods A logic model-informed mixed-methods program evaluation was conducted. The program comprised 10 sessions delivered over 12 days, totaling 36 contact hours, and included AI mental health foundations, emotion recognition, digital health, ethics, wearable technologies, interdisciplinary teamwork, and group-based prototype development. Evaluation data included pretest and posttest self-assessment questionnaires across 10 learning domains, open-ended qualitative responses, and descriptive review of five group-based AI mental health prototype proposals. Because pretest and posttest responses were de-identified and could not be matched, quantitative data were analyzed using descriptive statistics, mean change scores, and Welch independent-samples t tests. Qualitative responses were analyzed thematically. Results Fifty-four students participated; 50 completed the pretest and 33 completed the posttest. Most participants were female (70.0%), and 74.0% reported no prior AI experience. Scores improved significantly across all 10 learning domains (all P<.001). The largest improvements were observed in ability to design AI-enabled mental health solutions, emotion recognition and digital mental health, innovation and prototyping, STEM career readiness, and interdisciplinary collaboration. Qualitative findings showed that students valued practical AI tools, interdisciplinary communication, prototype development, STEM/AI career readiness, and human-centered clinical awareness. Five teams developed AI mental health proposals addressing grief, fatigue, suicide prevention, loneliness, and dementia caregiver burden. Conclusions This interdisciplinary, prototype-centered AI education program was associated with significant short-term improvements in students’ perceived AI knowledge, digital mental health readiness, interdisciplinary collaboration, innovation capacity, STEM career readiness, and solution-design ability. Trial registration: Not applicable.</p>