Lance B. Eliot, Stanford Center for Legal Informatics, has published The Dangers Of Miscalculating Human Behavior Via Robo-Judges. Here is the abstract.
There has been a trend lately of suggesting that robot judges (so-called robo-judges) will be able to determine the guilt of a defendant by merely using state-of-the-art sensory devices to watch and hear the accused. Besides the questionable aspects of whether this can or should be done, it also misportrays what AI-based judge-related systems might genuinely provide as an added value to adjudication.
Download the essay from SSRN at the link.
