Unfortunately, I haven't had the time to delve into AI, but one interesting conversation I had the other day we discussed one difference between (current) softw

Unfortunately, I haven't had the time to delve into AI, but one interesting conversation I had the other day we discussed one difference between (current) software and people is that a code duplication checker would miss a number of things a human could look at and recognize as obviously similar, if not identical. Applying things like Bayes can help (to an extent), but the question is, where is the line drawn? If we get a piece of software that can automatically refactor, does that count as AI?