Incident Tracker

Legal AI incidents on the record

165 verified incidents · as of 2026-10-07

2026-10-07United States (state)Sanctions
Kid International, LLC v. City National Bank of Florida
In Florida's Fourth District Court of Appeal, counsel for Kid International filed an AI-assisted brief citing a hallucinated case seven times and fabricated quotations from real authorities. The court affirmed on the merits, granted the opposing party's sanctions motion, and ordered the appellants and their counsel jointly and severally liable for the appellees' reasonable appellate attorney's fees.
2026-10-07United States (state)Dismissal
Tapscott v. Fariss
In the Court of Appeals of Iowa, a self-represented appellant's final brief in a protective-order appeal contained multiple hallucinated case citations. After two earlier briefs were struck and the appellant failed to answer the court's order to produce or explain the suspect citations, the court dismissed the appeal.
2026-10-07United States (state)Dismissal
Washington v. Justice
In Florida's Second District Court of Appeal, self-represented appellants filed a brief citing eleven cases the court could not find and likely generated by AI. After an unavailing show-cause response and a pattern of frivolous filings, the court dismissed the appeal and barred further pro se filings tied to the underlying case.
2026-10-02United States (state)Filing struck
Shannon v. Shannon, 322 Neb. 451 (2026)
In Nebraska, the state Supreme Court struck a self-represented divorce litigant's appellate brief for fictitious citations produced by his admitted AI use. Extending Prososki v. Regan to unrepresented parties, it held such briefs sanctionable, reviewed only for plain error, affirmed, and warned that future appeals may be dismissed outright.
2026-10-01OtherDismissal
P. M. M. c/ G. A., F. M. D. S.H. s/ cobro ordinario de sumas de dinero, Causa MO-14691-2021 (RS-722-2026)
In Argentina, a Buenos Aires provincial appeals chamber in Morón declared a fee appeal deserted, with costs, because it rested on a Supreme Court precedent that could not be found. The court said likely AI hallucination was the probable cause, warned counsel and asked the local bar association to remind members to verify AI-generated citations.
2026-09-29United States (federal)Warning
Hill v. Foundation Media LLC
In New York federal court, Judge Arun Subramanian ordered defence counsel to explain AI-affected quotations in a response brief, took no further action, and called the episode a wakeup call. Counsel's 1 October letter disclosed that Harvey, including updates referred to as Harvey 2.0, was used to prepare the response.
2026-09-28CanadaDismissal
Ahmed v. Canada (Citizenship and Immigration), 2026 FC 1202 (IMM-3886-26)
In Canada, the Federal Court refused a late filing in a refugee judicial review because the proposed memorandum cited five cases that could not be found and misstated others. Justice Lafrenière inferred unchecked AI use, called fake citations an abuse of process, and dismissed the leave application. No costs were ordered because the respondent sought none.
2026-09-25United States (federal)Sanctions
Aguilar v. The Crawford Group, Inc.
In Massachusetts federal court, Judge Angel Kelley sanctioned California-barred counsel under Rule 11 for fictitious and misattributed citations across three briefs. His firm must pay the Enterprise defendants' fees and costs, capped at $10,000, and the court revoked his pro hac vice admission. Local counsel was not sanctioned but was reminded of his supervisory duties.

SafeLegalAIrecords: safelegalai.com/tracker · CC BY 4.0 · not advice · no warranty