A judge in the Colombian port city of Cartagena has had artificial intelligence help with a legal decision. In a case involving the medical care of a child, he used the chatbot ChatGPT, said Judge Juan Manuel Padilla, according to media reports on Blue Radio. In addition, he used precedents for his judgement.

The current case was therefore about the question of whether the parents of an autistic child who only have an insufficient income are reimbursed by the health insurance for the costs of doctor’s appointments, therapy and transport. Padilla came to the conclusion that the cash register must cover these expenses.

Chatbots like ChatGPT, developed by the company “Open AI”, use artificial intelligence and tons of data from the Internet to generate answers to questions asked by human users or to write speeches, essays, cover letters, poems and any other text that sounds good should be as if a human had written them. In view of the technical advances in the field, this is getting better and better – and brings with it problems, for example in the field of education. ChatGPT recently made headlines when it was announced that the bot had passed exams at universities.

In the child’s case, Padilla asked the bot, according to the AFP news agency, among other things: “Are autistic minors exempt from paying fees for their therapies?” ChatGPT replied, “Yes, that’s correct. Under Colombia’s regulations, minors diagnosed with autism are exempt from paying fees for their therapies.”

While Padilla’s verdict itself didn’t draw much attention, the inclusion of the chatbot sparked debate about the use of artificial intelligence in legal matters. “It is certainly not responsible or ethical to use ChatGPT as the judge did in the decision in question,” said University of Rosario professor Juan David Gutierrez on Twitter. He asked the bot the same questions and got different answers. Gutierrez, an expert in the regulation and governance of artificial intelligence, called for judges to be trained urgently in “digital literacy.”

Octavio Tejeiro, a judge on Colombia’s Supreme Court, told British newspaper The Guardian that artificial intelligence is creating moral panic in the judiciary as people fear robots could replace judges. In his view, however, it will probably soon be accepted as a tool and become commonplace.

“The judicial system should make the best possible use of technology as a tool, but always respecting ethics and bearing in mind that the administrator of the judiciary is ultimately a human being,” quotes the Tejeiro newspaper. “It must be seen as a tool that helps the judge to improve his judgement. We must not allow the tool to become more important than the person.” He has not yet used ChatGPT himself, the judge explained, but is considering doing so in the future.

Padilla justified his use of the technology by saying it could make Colombia’s bloated legal system more efficient. ChatGPT does work that a secretary has previously done, “in an organized, simple and structured manner” that can “improve response times” in the judiciary.

Programs of this type could be useful to “facilitate the drafting of texts” but “not with the aim of replacing judges,” Padilla told AFP Blue Radio. He therefore emphasized that “by asking questions of the application, we do not stop being judges, thinking beings”.

The chatbot itself sees its new role in the judicial system more critically, as the “Guardian” writes. “Judges should not use ChatGPT when deciding legal cases […] It is not a substitute for the knowledge, expertise and judgment of a human judge,” he replied to a question from the newspaper – and added: “Journalists should exercise caution if they use ChatGPT-generated quotes in their articles.”

Sources: The Guardian, AFP