Regulating the Artificial Intelligence Economy: A Systematic Search and Mechanism‐based Review of Legal Institutions, Regulatory Mechanisms, Economic Trade‐offs, and Welfare Consequences

ABSTRACT

Artificial intelligence regulation changes technological risk as well as entry, learning, competition, work, and control. Legal and economic research nevertheless remains divided by instrument and discipline. This article combines a transparent systematic search with a mechanism‐based, purposive synthesis. The corpus contains 84 academic contributions and 20 primary legal and institutional documents, current through August 21, 2026. The search is reproducible, but the corpus is designed to cover mechanisms rather than estimate the prevalence of findings across all scholarship. The Institution‐Mechanism‐Incidence‐Welfare framework connects legal authority to behavior, strategic adaptation, and welfare. Four conclusions follow. Risk rules, assurance, liability, data rights, competition policy, and labor protections operate as complements. Fixed compliance and information costs can favor incumbents even when rules reduce direct harm. Evidence is strongest for narrower shocks involving privacy, workplace rollouts, and pricing technologies. It remains weakest for the aggregate welfare effects of comprehensive AI statutes. Efficiency claims are incomplete unless they report incidence and rights constraints separately. The research agenda therefore pairs safety with innovation, market structure, and distribution.