Principles alone cannot guarantee ethical AI
THE RADICALIZATION RISKS OF GPT-3 AND ADVANCED NEURAL LANGUAGE MODELS
Specific versus General Principles for Constitutional AI
Inverse Constitutional AI: Compressing Preferences into Principles
C3AI: Crafting and Evaluating Constitutions for Constitutional AI (very good read)
HUMAN FEEDBACK IS NOT GOLD STANDARD
Collective Constitutional AI: Aligning a Language Model with Public Input
Do large language models have a legal duty to tell the truth?