Action learning: exploring descriptive answers assessment method using artificial intelligence

Purpose This study explores action learning in the context of artificial intelligence (AI)-based assessment of descriptive answers, which can be institutionalized within organizations. It provides a process model and demonstrates its efficacy and validity. This research examines the long-term opportunities in AI usage for development and learning in organizations. Design/methodology/approach The methodology includes a human-in-the-loop process. The model consists of four sequential stages: AI-driven rubric generation, moderation of rubrics by faculty, AI-based grading and confirmation by faculty using random sampling. Findings The mean grades awarded by human professor and AI, showed no statistical difference. This validation is a step forward toward reinforcing the usage of AI tools in descriptive evaluation, through an action learning approach. Research limitations/implications This study fills a research gap in the area of AI-based assessment methods and serves as a foundation for future research. Practical implications Educational organizations (EO) face unpredictable technological changes that demand a high degree of novelty. This study validates a novel AI-based assessment of descriptive answers, the learning from which can improve faculty efficiency. It offers potential to develop further faculty competencies to transform digitally and improve the entire teaching-learning experience. The study has implications for the industry and society as well. Originality/value Literature on innovative AI adoption by faculty for grading, in context of the bigger picture i.e. subsequent transformation to other academic areas, is scarce.