Automatically Enriching Content for a Behavioral Health Learning Management System: a First Look

Abstract

Deep generative AI models have been evolving rapidly and are being applied to assist users in many domains. We consider an initial case study of applying this technology to semi-automated content enrichment for an online learning management system used by behavioral health professionals. Our goal is to optimize the use of domain expertise while also scaling the production efficiency of learning assets for users. We identify the possible opportunities for its use, discuss potential challenges and concerns. Finally, we provide prompt engineering strategies and initial quantitative results towards semi-automating one type of rime consuming editorial task, to gauge the feasibility of our approach. Results show that there is significant promise in using such an approach, and suggest that a larger, more rigorous study is required.