Whether we can actually safely develop AI or AGI is a much tougher question than whether that's the intent, unfortunately.