Our response draws on FLI’s ongoing research into autonomous agent governance, including analysis of real-world agentic incidents, technical governance frameworks, legal and institutional approaches to maintaining appropriate human control, and the intersection of AI security with broader societal resilience. Many of the topics we raise here were also present and supported in the industry working groups on agentic governance we are a member of at the Agentic AI Foundation and the World Economic Forum. We have organized our response according to the RFI’s numbered topic areas and have prioritized the questions NIST designated as highest priority.

With these deeper security and trustworthiness issues addressed, we believe the public and enterprises will more confidently and readily adopt such agentic productivity technologies.