Is Generative AI Mature Enough for Maturity Models? Insights from a Comparative Analysis

Abstract

Maturity models (MMs) serve as a basis to understand the improvement of quality. As an assessment tool, current capabilities are able to be recognized. With this, paths to higher levels, that yield better outcome, are made available for users. However, these MMs face challenges. Three of these challenges are considered in this study. The first challenge is market fluctuations, where MMs become outdated. Second, is the difficulty of finding an appropriate MM (assuming that an appropriate model even exists). Last, the creation process of an MM is, in general, significantly time and effort consuming. With the advent of generative-AI (GenAI), there seems to be potential in solving these problems. Since, in just an instant, GenAI can form an MM. This MM includes all the latest information known and is personalized, based on the prompt that has been given. This research sets out to discover the potential role that GenAI could play in the life cycle of an MM. To ground this, a literature review and comparative analysis were done. 17 interviews were conducted, where two selected human-created MMs were evaluated in contrast to two AI-generated variants of these models. All the models were compared in terms of quality. This study gives reasons to believe that AI-generated MM are on the same level, or even better, than human created ones. Additionally, evidence is shown that GenAI has a plurality of potential roles in the life cycle of an MM.