Generative artificial intelligence and entrepreneurial performance: the mediating and curvilinear role of innovation ambidexterity

Purpose Drawing on the resource-based view, the current research aims to explore how generative artificial intelligence (GenAI) adoption and GenAI capacities influence entrepreneurial performance (EP) in new ventures. It specifically examines how exploratory innovation (ERI) and exploitative innovation (EII) mediate these relationships and how their alignment or misalignment affects performance outcomes. Design/methodology/approach Based on a sample of 329 founders of entrepreneurial ventures, the study applies multiple linear regression and polynomial regression with response surface analysis to test a set of hypotheses derived from the theoretical framework. Findings The results reveal that both GenAI adoption and GenAI capacities have a positive effect on EP, mediated by ERI and EII. Furthermore, a balanced combination of ERI and EII strengthens performance outcomes, whereas incongruent combinations reduce them. Practical implications The study provides practical guidance for entrepreneurs and digital strategy professionals, emphasizing the importance of aligning innovation activities when leveraging GenAI for performance improvement. Originality/value This study extends the GenAI and entrepreneurship literature by identifying how innovation ambidexterity functions as a key mechanism through which GenAI contributes to EP. It also demonstrates the utility of a fit perspective using response surface analysis.