If they didn't you wouldn't need the operator, you'd have replaced all your programmers with no drawbacks by now. As long as we keep hiring humans that is all the evidence you need that these AI fails in ways humans don't.
Your argument "AI needs supervision, therefore it fails in different ways than its operator does" holds.
Your argument "AI fails in different ways than its operator, therefore the AI's intelligence is different in kind" doesn't hold.