A large-scale field trial on integrating generative AI into Pakistan’s justice system found that AI use resulted in an additional 1,848 cases being resolved per year, an increase of more than six per cent over the average, it emerged on Friday. A research paper titled “Courts of Tomorrow: Evidence from a Nationwide Rollout of Generative AI”, published on Tuesday, observed that “this increase does not appear to come at the expense of reduced quality.” The custom assistant, named JudgeGPT, was a chatbot based on OpenAI’s GPT-4 family of models, customised for the Pakistani context and beta-tested with the Federal Judicial Academy before deployment for use by 1,559 judges across 118 courts. The trial found that introducing the assistant alongside targeted training corresponded to an additional 1,848 cases resolved per year, a 6.3pc increase over the mean. Judges who received AI access with targeted training were more likely to adopt it, use it more intensively, and continue using it over time. The trial was conducted among “roughly half of the country’s trial judges and 80pc of district courts”, with 1,559 judges randomly divided into three groups: one with JudgeGPT and targeted training, one with the tool and generic training, and a control group with generic training only. Outcomes were measured via surveys and court records.