AI to take minor judicial decisions

Robo-judges and Robo-lawyers

In England and Wales, we are introducing a digital justice system that will allow citizens and businesses to go online to be directed to the most appropriate online pre-action portal or dispute resolution form. The Master of the Rolls, Sir Geoffrey Vos, has said in a speech that he believes that AI will be used within these symptoms to help people read and understand complex sets of rules and instructions, by limiting the material from which the answers to questions can be taken.

Sir Geoffrey Vos said that he predicts machines and Artificial Intelligence “may also, at some stage, be used to take some (at first, very minor) decisions” (Para 21). He explained that generative AI are not programmed to understand the full import of human question, nor are they fully aligned with human values and principles, meaning that humans are unlikely to ever accept personal decisions to be made by machines, such as those concerning the welfare of children. However, in commercial and compensation disputes, where fewer personal matters are involved, parties may have confidence in machine made decisions. Sir Geoffrey Vos explained that this is subject to certain controls, namely, (a) for the parties to know what decisions are being taken by judges and what by machines, and (b) for there always to be the option of an appeal to a human judge.

A crucial limitation for us all to understand is that ChatGPT might be missing crucial legal developments, as it has limited knowledge of the events that took place after 2021. Further, legal decision-making involves a range of factors beyond just predicting the outcome of a case, including strategic and ethical considerations, and client goals, that current generative AI is ill-equipped to consider. That is not to say, however, that AI would not be valuable when used in conjunction with human judgement and expertise.

If you are interested in reading Sir Geoffrey Vos’ full Law and Technology conference, you can access it here.