This PhD project uses a novel approach to model and study patterns of influence in human creativity and innovation by using artificial intelligence (AI) to study multi-modalities, i.e., language and image-based data. Additionally, we combine statistical learning methods with semantic enriched data representations in order to uncover the patterns and influences that affect the nature and originality as well as competitive performance of creative outcomes. Our approach is longitudinal, and the level of analysis is on both individual agents - e.g., artists, business organizations - and collectives - e.g., stylistic movements, (sub) genres. We integrate state-of-the-art AI methods that aim to answer questions regarding the origins of particular innovative developments for individual agents and collectives and their performance. In doing so, this project is directly relevant to a host of core management science questions pertaining to innovation management, strategy, marketing, and entrepreneurship.