Process of ensuring that an AI system's goals and behaviors are consistent with human values and ethics.
Generality: 790
