Agentic AI and Human Oversight
Published interviews, full transcripts, and questions to take into your work.
Published interviews, full transcripts, and questions to take into your work.
When AI systems take actions, organizations need to understand their permissions, boundaries, and routes back to a person. The Signal Room's agentic AI interviews examine those issues through healthcare workflows, business operations, data security, and patient access.
Kaled Alhanafi discusses why agents need to know when to stop. Travis Garland explores data leaks and governance, and Pranava Adduri examines security risks associated with AI agents. Ethan Seib considers human oversight in business automation. Jakub Grajcar discusses outpatient access, providing a setting in which to examine voice automation and the patient experience.
Together, these interviews give listeners concrete questions to bring to a proposed deployment: What can the system access? Which actions can it take? When should it stop or escalate? How will a person understand what happened? Each guest describes a particular perspective or application. Follow the full transcripts to distinguish those experiences from general claims about all agents, and explore the data and security library for related foundations.
AI search and discovery · Leadership and work · Agentic AI and oversight · All episodes
For cross-sector AI and technology conversations, visit our companion show, The True Up. PR representatives can propose a focused interview; companies seeking a disclosed commercial partnership can contact the sponsorship team.