Emelia Probasco, Kathleen Curlee, and Autumn Toney authored a report titled “Trust Issues: Discrepancies in Trustworthy AI Keywords Use in Policy and Research,” which was showcased in the FAccT ’24: Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency.
The authors discuss how definitions of AI ethics by governments, practitioners, and researchers impact AI models and systems. They emphasize the need for policy and technical convergence to establish international standards for trustworthy AI. Analyzing over 322,000 scientific papers and national documents from Australia, Canada, Japan, the United Kingdom, and the United States, they found both similarities and relevant differences in definitions and frameworks.