TY - JOUR AU - Reyes-Portillo, Jazmin A AU - So, Amy AU - McAlister, Kelsey AU - Nicodemus, Christine AU - Golden, Ashleigh AU - Jacobson, Colleen AU - Huberty, Jennifer PY - 2025 DA - 2025/7/28 TI - Generative AI–Powered Mental Wellness Chatbot for College Student Mental Wellness: Open Trial JO - JMIR Form Res SP - e71923 VL - 9 KW - artificial intelligence KW - chatbots KW - anxiety KW - depression KW - college students AB - Background: Colleges have turned to digital mental health interventions to meet the increasing mental health treatment needs of their students. Among these, chatbots stand out as artificial intelligence–driven tools capable of engaging in human-like conversations that have demonstrated some effectiveness in reducing depression and anxiety symptoms. Objective: This study aimed to assess the feasibility and acceptability of using Wayhaven, an artificial intelligence chatbot, among college students with elevated depression or anxiety symptoms. We also aimed to examine the preliminary effectiveness of Wayhaven in improving symptoms of anxiety and depression, hopelessness, agency, and self-efficacy among students. Methods: Participants were 50 racially and ethnically diverse college students with elevated depression or anxiety symptoms (n=45, 80% female; mean age 22.12, SD 4.42 years). Students were asked to use Wayhaven over the course of 1 week and completed assessments at preintervention, after 1 session, and 1 week. Results: Wayhaven use was associated with a significant decrease in depression (β=−1.62; P<.001), anxiety (β=−2.15; P<.001), and hopelessness (β=−.64; P<.001) and a significant increase in agency (β=.64; P=.32), self-efficacy (β=.53; P=.02), and well-being (t40=2.90; P=.006; d=0.45) across the study period. Most students also reported being satisfied with Wayhaven and it being a tool they would recommend to their peers. Conclusions: Findings suggest that Wayhaven may be a viable mental wellness resource for diverse students with elevated depression or anxiety symptoms. SN - 2561-326X UR - https://formative.jmir.org/2025/1/e71923 UR - https://doi.org/10.2196/71923 DO - 10.2196/71923 ID - info:doi/10.2196/71923 ER -