I would love to see what would come out of such a NN trained on a set of chess games. Would it "make the next valid move"? Would it appear to behave tacticall

I would love to see what would come out of such a NN trained on a set of chess games. Would it "make the next valid move"? Would it appear to behave tactically? Strategically?