Year
2023
File Attachment
finalpaper_411_0524100342.pdf360.39 KB
Abstract
Artificial Intelligence (AI) based applications are on the cusp of offering the public and private sectors
tools of tremendous potential that will likely transform the world in much the same way that previous
technological revolutions have. To harness these tools, a vast and rapidly increasing assortment of
ethical AI guidelines and principles are being developed to manage the myriad of risks these tools
pose. Documents, frameworks, standards, and regulations have been developed by private industry,
research institutions, governments and nongovernmental organizations, international standards
bodies and more. This work details representative approaches to ethically developing, managing, and
operating current AI technologies. The various approaches are analyzed for commonality, divergence,
and implementation strategy to help develop approaches for managing AI tools in the nuclear
landscape. This work will support the creation of a common vocabulary and a deeper understanding
of the ethical/ responsible AI landscape and advance the considerations required for structured
frameworks in the international safeguards domain.