The Alignment Challenge
From Opiniowiki
The key challenge is alignment. How do we ensure that an AI agent, when given a complex goal, pursues it in a way that aligns with human values and avoids unintended, harmful side effects? This is a major area of research.
