I believe the bar for publication should be "can this be understood by (expert) humans". If that's not the case, it may even be true, but who cares? It's not improving knowledge, which is by definition human understanding.