How can AIs know what we want if *we* don't even know? (with Geoffrey Irving)
Clearer Thinking with Spencer Greenberg
00:00
Navigating Constitutional AI
This chapter explores the establishment of rules for AI systems to ensure alignment with human preferences, focusing on collaborative efforts by expert committees. It discusses the challenges in defining boundaries across various domains, such as medical inquiries, and emphasizes the importance of balancing comprehensive guidelines with adaptability to cultural and ethical standards. The conversation also addresses the complexities of conflicting rules and the need for safe defaults in AI behavior, highlighting the necessity for collaboration among organizations to develop ethical AI technologies.
Transcript
Play full episode
