Recently, a judge in Colombia made headlines all around the world when he announced that he had used the AI chatbot ChatGPT in preparing a ruling on a case concerning children’s medical rights. This isn’t the first time ChatGPT has been used to assist with legal decisions – it has been used in a number of countries, including the United States and India – but it is the first time it has been used in Colombia, and the case in question is a particularly sensitive one.
At its core, ChatGPT is a software program designed to generate and respond to complex conversations ‘gleaned’ from a digital archive of an already existing corpus of related conversations. It works by breaking down and understanding the conversation to answer questions by ‘mapping’ existing conversations. This allows it to generate an answer or response based on its understanding of the given conversation, and it can learn new conversational topics over time.
So how did ChatGPT help the judge in Colombia? In this case, ChatGPT helped by providing the judge with more contextual and detailed information on the case. The judge was presented with insights and opinions from past cases that had similar facts and circumstances, which in turn enabled him to make a more informed decision. In addition, ChatGPT was also able to ask relevant questions, acting as an aid in providing evidence to build a case.
This decision caused quite a stir as people on both sides of the argument criticized the judge for relying heavily on an AI-based system to make a decision in such a sensitive case. However, supporters of the use of ChatGPT argue that it can actually be beneficial to have AI-driven tools providing the court with additional context or evidence that might have been overlooked by the judge in a time-strapped environment.
Ultimately, this decision highlights the increasing sophistication and potential use of AI-based tools to support courts in their decisions. While it will be interesting to see how this particular case turns out, this is a reminder that we must also be aware of the potential risks associated with relying heavily on AI to make decisions.