Here is a new 'KQRBNP vs krrbn' chess problem generated autonomously by the prototype computer program, Chesthetica, using the Digital Synaptic Neural Substrate (DSNS) computational creativity approach. It doesn't use endgame tablebases, deep learning or any kind of traditional AI. Chesthetica is able to generate three-movers, four-moves, five-movers and study-like constructs and also compose problems using specific pieces types fed into it (e.g. compose something original using a knight versus three pawns). Read more about it on ChessBase. This position contains a total of 11 pieces. The largest endgame tablebase in existence today is for 7 pieces (containing over 500 trillion positions anyway) which means the problem could not have been taken from it regardless.
