Moral hazard problem

1. Positive AI Alignment vs Negative AI Alignment (Prompt Engineering Secrets)