🔍 Read the full analysis: AI And Fake Court Precedents In Korea: What Judges Reported on ThorstenMeyerAI.com
Get hardware and tech essentials 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 reported that about 80% of responding civil, family and administrative judges had encountered false or nonexistent precedent citations. The survey covered 360 judges, and its results point to concerns about filings by both litigants and lawyers, as well as calls for formal rules on verification and AI disclosure.
South Korea’s National Court Administration reported on November 2 that about 80% of responding judges handling civil, family and administrative cases had encountered false or nonexistent precedent citations in court filings. The survey of 360 judges also found that respondents attributed such material to both litigants and lawyers, prompting calls for formal measures including fines, citation checks and disclosure of AI use.
The survey was conducted from October 6 to 16, 2026, among judges at courts of all levels. The administration said AI-generated hallucinations can produce false citations that place an unnecessary verification burden on opposing parties and may slow trials. The survey records what judges said they had encountered; it does not establish the share of all filings containing fabricated material.
Among the judges who reported encountering false citations, 196 (80.7%) identified precedents that did not exist, while 159 (65.4%) identified real precedents cited with false content. Respondents could select more than one category. The administration also recorded reports of false or nonexistent statutes and legal publications, though at lower rates.
The denominator behind the widely reported “about 80%” matters. That figure applies to 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 itself print that all-respondent percentage.
Why False Citations Strain Court Filings
Judges must be able to check the authorities cited in a filing. A nonexistent case or an altered account of a real ruling can require the other side and the court to spend time verifying material instead of addressing the legal dispute. The court administration said this can burden opposing parties and interfere with the speedy progress of trials.
The findings also show why the issue is not limited to people representing themselves. Asked who had submitted false material, 164 judges (69.5%) identified the parties themselves, while 136 (57.6%) identified lawyers acting as representatives or defense counsel. Those figures describe judges’ reported observations, not a verified count of individual filings or practitioners.
Most respondents supported an institutional response: 91.7% said one was needed, and 64.8% described it as very necessary. The most frequently selected options included a legal basis for fines, a duty to verify citations and a duty to disclose AI use. The results do not show that any of these proposals has already become law.
Products Worth Considering
Apple MacBook Pro 16.2-inch with M5 Pro, 18-Core CPU, 20-Core GPU, 48GB Memory, 1TB SSD, Wi-Fi 7, Silver
Ledger Nano Gen5 Crypto Wallet, Black
Ledger Flex Crypto Wallet with 2.8-Inch E Ink Touchscreen
As an affiliate, we earn on qualifying purchases.
How the Korean Survey Was Framed
The National Court Administration is the administrative arm of South Korea’s Supreme Court. Its Judicial Support Office issued a four-page release 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 could lead to citations of false or nonexistent precedents.
The release gave examples of material judges had encountered: fabricated judgments formatted like Supreme Court decisions; unfamiliar reasoning paired with a Supreme Court case number that could not be found; real rulings cited with altered content or in a way that reversed their meaning; and text from AI conversations or litigation-strategy notes pasted into filings. These are examples reported by the administration, not a separate audit of court documents.
There is also a translation issue in some coverage. An English-language Yonhap report republished by The Korea Herald described 164 judges as saying they themselves had cited false precedents. The Korean-language wording identifies the litigants themselves as the source. The distinction matters: the survey did not report that judges admitted submitting fabricated citations.
What the Survey Cannot Establish
The administration did not disclose how many judges were invited, how the questions were worded or whether respondents were representative of the judiciary as a whole. Because the survey asks judges what they had seen, its results do not measure the prevalence of fabricated citations across all filings, courts or cases.
The published material also does not verify each reported incident or identify the people, law firms or AI systems involved. The available results describe selected responses, with multiple answers allowed for citation types and sources. The survey therefore cannot show how often one filing contained false material, or how many filings were produced using AI.
It remains unclear which proposals will be taken forward, what standards would govern any duty to verify or disclose AI use, and how fines would be applied. The administration’s stated plan to propose changes is not evidence that legislation has been introduced or enacted.
Proposed Procedural Law Changes
The National Court Administration plans to propose changes to South Korea’s Civil and Criminal Procedure Acts that would provide a legal basis for fines related to false citations. The survey also records support for citation-verification and AI-disclosure duties, but the release does not specify whether those measures will be included in proposed legislation.
The next concrete developments to watch are whether the administration submits draft amendments, how lawmakers define prohibited conduct and what safeguards or procedures accompany any penalties. Until those steps are confirmed, the survey’s findings are evidence of judges’ reported experiences and preferences—not new requirements for litigants or lawyers.
Key Questions
What did South Korea’s judges report?
The court administration said about 80% of responding civil, family and administrative judges had encountered false or nonexistent precedent citations. The figure applies to that group, not to every judge or every court filing.
Did the survey show that 80% of all filings contain fake citations?
No. It asked judges whether they had encountered such citations; it did not measure the share of filings containing them. The administration did not publish a count of affected documents.
Who did judges say submitted the false material?
Respondents identified both litigants and lawyers. The survey reported that 164 judges (69.5%) selected the parties themselves and 136 (57.6%) selected lawyers. These are survey responses, not confirmed counts of individual cases.
What action did judges support?
91.7% of respondents called for an institutional response. Frequently selected measures included a legal basis for fines, a duty to verify citations and a duty to disclose AI use. The results do not mean those measures are already in force.
What happens next?
The National Court Administration said it plans to propose amendments to the Civil and Criminal Procedure Acts to allow fines. The timing, final contents and legislative outcome remain unclear.
Source: ThorstenMeyerAI.com
