Firm Preferences over International AI Regulatory Governance

• In Event: The Global Governance of Artificial Intelligence

Abstract

What do firm managers want with regards to AI regulation? The study of the politics of AI regulation is nascent, yet the call to action from both tech and government leaders is urgent. This study implements survey and online field experiments to explore how firm managers assess different AI regulatory initiatives. In particular, we study how exposure to information about specific potential threats from AI affects preferences over international regulation. We deploy a conjoint survey in which subjects are firms primed on concerns about AI violations of privacy, algorithmic bias, systematic error, or discrimination based on AI-revealed personal attributes, then presented with hypothetical international regulatory institutions varied along the dimensions of precision, obligation, and delegation. Our study fills a hole in our understanding of the politics of AI by studying firm elites and comparing their attitudes and behavior with citizens.

Authors

• Terrence Chapman, University of Texas at Austin

• Daniel L. Nielson, University of Texas at Austin

• Huimin Li, University of Texas at Austin