South Korea’s National Court Administration is reviewing legislative and procedural changes to address the submission of falsified legal documents created by artificial intelligence. The move follows instances where legal professionals have cited non-existent cases and fabricated precedents generated by AI language models in official court filings.
The judiciary’s administrative body announced it is exploring amendments to laws such as the Civil Procedure Act. The proposed revisions aim to establish clear penalties for submitting false evidence, including the imposition of fines on legal representatives who file documents containing AI-generated fabrications. The objective of the review is to safeguard the integrity of legal proceedings and ensure the reliability of evidence presented in court.
The issue stems from the use of generative AI tools, which can produce convincing but factually incorrect information, a phenomenon often referred to as “hallucination.” In the legal context, this has resulted in lawyers unknowingly including citations to court cases that do not exist in their briefs and arguments.
In response to this emerging challenge, the Korean Bar Association has also taken action. The association distributed ethical guidelines to its members concerning the use of AI in legal practice. These guidelines emphasize the professional responsibility of lawyers to verify all information, including legal precedents and case details, before submitting them to the court. The association underscored that the ultimate responsibility for the accuracy of filed documents rests with the human lawyer, not the AI tool.
The situation in South Korea reflects a global concern among legal professionals and judiciaries. Similar incidents have been reported in other countries, prompting discussions on the need for updated regulations. In a widely reported 2023 case in the United States, two lawyers were sanctioned by a federal judge in New York for submitting a legal brief that cited multiple non-existent judicial decisions created by an AI chatbot.
The review by the National Court Administration is currently in progress. Officials have stated the goal is to develop a framework that holds legal practitioners accountable for the information they present, thereby upholding the standards of the country’s justice system as technology continues to evolve. No final decisions on specific legislative changes or timelines for implementation have been announced.
