Lawyers in KwaZulu-Natal, South Africa, relied on AI-generated case studies in a High Court case in Pietermaritzburg, resulting in the court's dismissal of the plaintiff's appeal. The judge discovered that the AI chatbot, ChatGPT, had fabricated several cited cases, leading to a ruling against the plaintiff, Philani Godfrey Mavundla, a suspended mayor. The court criticized the lawyers for "false trust in the veracity of AI-generated legal research" and for failing to verify the information. This incident is one of at least three similar cases in South Africa involving AI misuse in legal proceedings.

The incident prompted calls for greater ethical guidelines and responsible AI usage within the legal profession. Tayla Pinto, an AI law specialist, highlights the growing problem of lawyers not using generative AI responsibly. Mbekezeli Benjamin, a human rights lawyer, stresses the need for clear guidelines and potential code of conduct amendments to address the issue, suggesting penalties for inappropriate AI use could include fines or removal from the professional register. The Legal Practice Council (LPC) acknowledges the issue but currently believes existing regulations suffice, though the debate continues.

Impact Statement: The case underscores the need for ethical guidelines and responsible use of AI in legal research, highlighting potential risks of relying on AI-generated information without proper verification.