#12: Michael K. Cohen on Regulating Advanced Artificial Agents
Center for AI Policy Podcast
00:00
Navigating AI Control and Oversight
This chapter explores the challenges of evaluating actions in advanced AI systems through a pessimistic lens, particularly focusing on the alignment and control of superintelligent agents. It discusses scalable oversight methods and contrasts different learning paradigms to highlight their implications for ensuring human oversight.
Play episode from 02:00
Transcript
