Daron Acemoglu’s essay Harms of AI advocates broad state regulation of artificial intelligence (AI) to mitigate social and economic risks. This article presents a systematic critique of this approach, grounded in Friedrich Hayek’s theory of spontaneous order, Milton Friedman’s principle of economic scarcity, and Leonard Read’s metaphor of decentralized cooperation. It argues that attempting to centrally regulate such a complex phenomenon as AI falls into the “fatal conceit” of disregarding the dynamics of dispersed knowledge and spontaneous coordination among market agents. Through analysis of AI’s global production chain and
its multiple interdependent components, we demonstrate that free markets and competition constitute more effective instruments for risk mitigation than centralized regulations, which tend to inhibit innovation and create economic distortions.