A computer can never be held accountable, therefore a computer must never make a management decision.
Except this should really be “any kind of choice or decision” in general, not just management ones.
This means that AI should never be allowed to make any kind of decision or be able to employ any kind of choice, including what code to create and how the program gets constructed, because it can never be held accountable.
Comments
For me, I take a page out of IBM’s book:
Except this should really be “any kind of choice or decision” in general, not just management ones.
This means that AI should never be allowed to make any kind of decision or be able to employ any kind of choice, including what code to create and how the program gets constructed, because it can never be held accountable.