🔍 Read the full analysis: Why Fake Precedents Matter In Korean Courts As AI Use Grows on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
South Korea’s National Court Administration published a survey on November 2, 2026, in which judges reported encountering fabricated or distorted legal citations, including nonexistent precedents and altered descriptions of real rulings. The results measure judges’ reported experience, not the share of court filings containing false material; the administration is considering procedural law changes, but no amendments have been adopted.
South Korea’s National Court Administration published its first survey of judges’ experience with fake or distorted citations on November 2, reporting that about 80% of responding judges handling civil, family and administrative cases had encountered them. The findings have prompted calls for fines, citation-checking duties and disclosure of AI use, as the administration considers changes to procedural laws.
360 judges from courts at all levels answered the survey between October 6 and 16, according to the court administration’s release. The results describe what judges said they had seen; they do not establish what proportion of filings contain inaccurate citations. The administration did not state how many judges were invited or provide the wording of the questions.
Among judges who reported encountering false or nonexistent precedents, 196 (80.7%) selected nonexistent rulings, while 159 (65.4%) selected real precedents cited with false content. Respondents could select more than one category. Other reported categories included false or nonexistent statutes and academic papers or literature.
The denominator matters: the roughly four-in-five figure concerns responding judges handling civil, family and administrative cases. Kyunghyang Shinmun reported that 243 of all 360 respondents (67.5%) had encountered fake citations. The court administration’s release did not print that overall figure. Judges also reported submissions from both unrepresented litigants and lawyers: 164 judges (69.5%) named parties themselves, and 136 (57.6%) named lawyers acting as representatives or defense counsel.
Why False Citations Burden Trials
A citation to a nonexistent judgment can force an opposing party and a judge to spend time checking whether the authority exists and whether it says what the filing claims. The court administration said false citations create unnecessary verification work for opposing parties and may slow the progress of trials.
The survey also points to a problem that cannot be reduced to whether a litigant used AI. The administration’s examples included fabricated judgments formatted to resemble Supreme Court decisions, real rulings described inaccurately, and text from AI conversations or litigation-strategy notes copied into filings without editing. The survey identifies judges’ observations, however; it does not establish which examples were generated by AI or how often AI was involved.
The responses suggest judges want rules that apply beyond informal guidance. The administration reported that 91.7% supported an institutional response. The most selected proposals included a legal basis for fines, a duty to verify cited precedents and a duty to disclose AI use. Those preferences do not show that any of these measures has been enacted.
Products Worth Considering
Logitech MX Brio Ultra HD 4K Webcam, Graphite
Bose SoundLink Flex Bluetooth Speaker (2nd Gen) – Portable Outdoor Speaker with Hi‑Fi Audio, Waterproof and Dustproof, USB‑C, Up to 12 Hours
Amazon Kindle Paperwhite (2024), 16 GB, 7-inch Display, Black
As an affiliate, we earn on qualifying purchases.
What the Survey Actually Measured
The results came from a four-page Korean-language release by the National Court Administration’s Judicial Support Office, titled “Results of the Survey on Citations of AI-Generated False Precedents and Similar Material.” The administration said the survey was prompted by concerns that AI hallucinations can produce false or nonexistent legal authorities. It described the consequences as extra checking work and a possible impediment to speedy trials.
The figures should be read as a measure of judges’ reported experience, not an audit of court documents or a count of fabricated citations. The source release says the percentages use the respondents to each question as their base and that multiple answers were allowed for the categories of false material. It does not provide enough information to calculate the prevalence of false citations in filings across South Korean courts.
One English-language account of the results illustrates why original wording matters. The Korea Herald republished an English Yonhap report saying that 164 judges had said they themselves cited false precedents. The Korean release instead identifies the parties themselves as a source named by those judges. The survey does not say that the 164 judges admitted citing false cases.
How Common Are False Filings?
The survey does not show the number or share of filings containing false citations. It also does not say how many judges were invited, provide the question wording, or establish whether the cited material was produced with AI in each instance. The published findings are based on judges’ reports, not an independent examination of court records.
It is also unclear what safeguards would be adopted, what conduct would qualify for a fine, or how an AI-use disclosure duty would work. The survey records judges’ preferences; it does not show agreement among lawmakers or a final policy decision.
Proposed Changes to Court Rules
The National Court Administration plans to propose changes to South Korea’s Civil and Criminal Procedure Acts that would allow fines for false citations, according to the source report. The survey release and reported plan do not mean that the laws have already changed. A proposal would still need to proceed through the relevant legislative process.
For now, the next substantive developments are whether the administration publishes or submits specific language, and whether lawmakers consider fines alongside duties to verify citations or disclose AI use. Until then, the survey’s clearest finding remains a report of judges’ experiences and their support for a formal response—not a measured rate of AI-generated errors in court filings.
Key Questions
What did South Korea’s survey find?
The National Court Administration said about 80% of responding judges handling civil, family and administrative cases had encountered fake or distorted citations. The figure concerns judges’ reported experience, not the percentage of court filings that contain errors.
Does the survey show that eight in ten court filings contain fake citations?
No. It reports what judges said they had encountered. It does not audit filings or estimate the share of documents containing false citations.
Did judges say they themselves cited fake cases?
No. The Korean release says 164 judges named the litigants themselves as a source of false material. An English-language report mistranslated this point; the survey does not say those judges admitted citing fake cases.
What rules could South Korea consider?
Judges most often supported a legal basis for fines, a duty to verify cited precedents and a duty to disclose AI use. The administration plans to propose procedural law changes to allow fines, but no such changes have been adopted based on the information published with the survey.
Source: ThorstenMeyerAI.com
