AI Literacy, Human-AI Interaction, and Generative-AI Decision Support for Entrepreneurial Opportunity Seeking: A Convergent Mixed-Methods Study

Abstract

 Generative artificial intelligence (AI) systems are increasingly used as interactive information and decision-support tools, but their value is associated with users’ capacity to formulate requests, interpret outputs, verify information, and exercise ethical judgment. This study examines AI literacy as a multidimensional user capability associated with AI-assisted entrepreneurial opportunity seeking. A convergent mixed-methods design combined survey data from 57 Bachelor of Science in Entrepreneurship students with qualitative interviews and focus-group responses from 14 participants. AI literacy was measured through awareness, application skills, evaluation of AI outputs, and ethical use, while AI-assisted opportunity seeking covered opportunity discovery, evaluation, and development. The four literacy dimensions jointly significantly predicted AI-assisted opportunity seeking in the regression model: R² = .506; adjusted R² = .468; F(4,52) = 13.315, p < .001. AI awareness (B = .302, p = .016), application skills (B = .313, p = .047), and ethical use (B = .244, p = .016) were significant positive predictors; evaluation of AI outputs was not significant (B = -.041, p = .751). Qualitative data confirmed themes of prompting difficulty, unreliable outputs, verification problems, internet-access constraints, weak contextual fit, and overdependence. These challenges were associated with lower-quality human-AI interaction. The study contributes to computer information systems research by identifying a gap between operational AI use and information-quality evaluation. It proposes design and governance implications for prompt scaffolding, source provenance, verification workflows, contextual data integration, and human oversight in AI-enabled entrepreneurial decision support.