Reaction
A plan to link legal theory and AI reward design
A new cohort member says they want more rigorous, interdisciplinary ways to write rules and design rewards for AI systems.
TLDR
A new cohort member says legal theory overlaps with reward design in reinforcement learning and plans to explore better ways to write AI rules and design rewards. Another poster links to a paper they describe as their own on how law can improve rule-based alignment.
Combined views
889
3 Sources, first seen ago
