The field trying to ensure AI doesn’t cause us harm. Think of it as the digital equivalent of childproofing your house, except the child is potentially smarter than you and might figure out how to circumvent all your precautions.
Synonyms:
AI Alignment