POLICY BRIEF - AI Governance Needs Sociotechnical Expertise - Why the Humanities and

Technical de- sign, social practices and cultural norms, the context a system is integrated in, and who de- signed and operates it all impact the performance, failure, benefits, and harms of an AI sys- tem.2 As the National Institute of Standards and Technology’s (NIST) AI Risk Management Framework (RMF), which was designed to provide guidance to organizations developing, deploying, or using AI sy. [...] Humanities experts, social scientists, and user experience (UX) researchers and designers offer a tangible starting point for expanding the sociotechnical analysis of generative AI and addressing its technical and so- cial dimensions simultaneously.17 For example, University of Washington researchers applied the method of autoethnography to track their members’ use of generative AI and associated. [...] Federal hiring efforts related to the National AI Talent Surge and the Office of Management and Budget’s (OMB) memo on agency use of AI must in- vest in humanities and social science experts.32 Humanities and social science experts should be a focus of hiring efforts aimed at addressing the many issues with and potential opportunities of AI systems, as well as for roles facilitating meaningful pub. [...] Humanities and social science methods and expertise should be included in the development of all guidelines and standards for AI assessment, re- search and development (R&D), and policy. [...] Auditing and Social Justice: Chicago: University of Chicago Lessons from the History of Audit Press, 2016; Ronald Kline, “How Studies,” In Proceedings of the 1st Disunity Matters to the History of ACM Conference on Equity and Cybernetics in the Human Sciences Access in Algorithms, Mechanisms, and in the United States,” 1940–80 Optimization, pp.