How does information about AI regulation affect managers' choices?

Benjamin Cedric Larsen coauthors blogpost at Brookings

Benjamin Cedric Larsen

With coauthors Yong Suk Lee, Mariano-Florentino (Tino) Cuéllar, and Michael Webb, we set out to answer "how information about AI regulation affects managers' choices."

We find that information about varying kinds of AI regulation increases managers' perception of the importance of safety, privacy, bias/discrimination, and transparency issues related to AI. However, there is a tradeoff; regulation information reduces managers' intent to adopt AI technologies.

Such tradeoffs are likely to be associated with industry-specific characteristics, which holds implications for how new and intended AI regulations could affect varying industries differently.

We recommend that businesses embrace new managerial standards and practices that detail AI liability under varying circumstances, even before it is regulatorily prescribed. More robust internal audits and third-party examinations would provide more information for managers, reduce managerial uncertainty, and aid the development of AI products and services that are subject to higher ethical as well as legal and policy standards.