Purpose: This paper aims to examine the feasibility of artificial intelligence (AI) performing as chief executive officer (CEO) in organizations. Design/methodology/approach: The authors followed an explorative research design – classic grounded theory methodology. The authors conducted face-to-face interviews with 27 participants that were selected according to theoretical sampling. The sample consisted of academics from the fields of AI, philosophy and management; experts and artists performing in the field of AI and professionals from the business world. Findings: As a result of the grounded theory process “The Vizier-Shah Theory” emerged. The theory consisted of five theoretical categories: narrow AI, hard problems, debates, solutions and AI-CEO. The category “AI as a CEO” introduces four futuristic AI-CEO models. Originality/value: This study introduces an original theory that explains the evolution process of narrow AI to AI-CEO. The theory handles the issue from an interdisciplinary perspective by following an exploratory research design – classic grounded theory and provides insights for future research.