There's also a plethora of writings showing that using only using humans in court _will and has_ go wrong. There's probably a plethora of writings showing that

There's also a plethora of writings showing that using only using humans in court _will and has_ go wrong. There's probably a plethora of writings showing that using both humans and AI _will_ go wrong. The question isn't _will_ they go wrong (they will) it's a question of will they do it less.