My heuristic is that it's usually faster to catch-up to something that already exists than it is to invent something new. Given sufficient effort Apple should be able to asymptotically approach Google's capabilities. But considering the other two points I mentioned, it's not a given that they will.