Regulation of Artificial Intelligence: A Systematic Literature Review
Regulation of Artificial Intelligence: A Systematic Literature Review

Summary/Abstract: Objectives: This study aims to examine the different approaches used to regulate artificial intelligence through a systematic and bibliometric analysis of the literature. Prior Work: Governments and industry concur that there is an urgent need for artificial intelligence (AI) regulation. The European Union, the United States of America and China have already pioneered in the regulation of AI with different regulatory approaches. However, there remains no clearly defined method of how AI can be regulated comprehensively. Approach: A total of 38 articles published between 2019-2024 extracted from the Scopus and Clarivate Web of Science databases were selected for analysis following the PRISMA protocol. The review combined bibliometric and systematic review. Results: The study established risk-based, management-based, rights-based, systemic market safety, tort liability, single unified law and ethics as major approaches used to regulate AI. Economic and political concerns remains barriers to effective regulation as governments and jurisdictions aim to become centers of innovation. Implications: A universal and unified regulation of AI which extends from ethical standards enforced by professional bodies to actual enforceable law is required to avoid law lacunas. Value: Major AI regulatory approaches were examined, and trends and gaps in the regulation of AI literature were identified.

• Details
• Contents