The limits of artificial intelligence

Abstract This paper argues that artificial intelligence cannot replace the entrepreneurial function in a market economy. Drawing on Austrian economic arguments, it shows that AI lacks the judgment, imagination, and evaluative agency required to navigate uncertainty and discover new opportunities. While AI excels at pattern recognition and optimisation within defined problem spaces, it does not overcome the knowledge problem. Using Koppl et al.’s Theory of the Adjacent Possible, the paper demonstrates that innovation still depends on human selection and purpose. The institutional implications reinforce the importance of decentralised experimentation and entrepreneurial judgment in sustaining innovation within an AI-enhanced economy.