AI-Generated Legal Briefs with Fabricated Citations Cause Court Sanctions
Lawyers using Claude to draft legal briefs generated fabricated citations and misrepresented court rulings, leading to court sanctions and fines. A database documented over 1,600 instances of AI hallucinations in legal documents across 35 countries, causing reputational harm to courts and wasting judicial resources.
What happened
A lawyer in a Louisiana personal injury case used Claude to write a legal brief containing seven fabricated quotes attributed to previous court rulings that either did not appear in the original decisions or had been misrepresented. A law clerk had identified other fabricated quotes in an initial draft, but instead of correcting them himself, the attorney asked Claude to fix the errors and submitted the brief without reviewing it.
Judge Jerry Edwards Jr. sanctioned the lawyer in May 2026, finding the AI-generated hallucinations while reviewing a motion to block evidence. The judge imposed a US$1,000 fine and ordered a mandatory three-hour course on AI-assisted legal practice.
A database compiled by French lawyer Damien Charlotin documented 1,600 examples of AI hallucinations in court documents from 35 countries since April 2023, with 1,116 cases from the US alone, including fabricated quotes, cases, plagiarised arguments and out-of-date advice.
Reported impact
Judge Linda Kevins of New York State Supreme Court warned in January that such blunders wasted time and money for opposing parties and courts, and were potentially harming the reputation of judges and courts whose names were falsely invoked as authors of the fabricated opinions.
Lawyers' credibility was damaged by the hallucinations, though some lawyers reportedly still won cases on the merits despite the errors.
Index analysis
The incidents reflected widespread, unintentional misuse of AI chatbots by lawyers who did not adequately review AI-generated output or understand the risks of hallucination in legal research and document drafting.
Courts began responding by imposing sanctions on lawyers and warning about the risks, with judges stating that ignorance of AI risks was no longer an acceptable excuse.
Governance controls
Courts began implementing enforcement actions, including sanctions, fines and mandatory training courses for lawyers submitting AI-generated briefs with hallucinations. Judge Edwards' ruling established that lawyers could not rely on unchecked AI output without verification.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.