Sanction teardown · D. Utah, USA · 2026-07-30
Carey v. Breakell et al.
What happened
In D. Utah, USA, a filing relied on an unnamed/unconfirmed AI tool to help draft legal argument. The court identified the following problems with the citations in that filing:
-
Misrepresented (Case Law)The Opposition miscited a real Tenth Circuit decision (Huff v. Reeves) as having been issued in 2022; the court noted the miscitation in its review.
-
Fabricated (Case Law)Opposition cited 'Larsen v. Davis Cnty. Sch. Dist., 2017 UT App 27, ¶¶ 14–16, 392 P.3d 1008' which the court could not locate and treated as a non-existent opinion.
-
Fabricated (Case Law)Opposition cited 'Dale v. Bennett, 921 P.2d 466, 469 (Utah 1996)' which the court was unable to locate and identified as a likely fake opinion.
Which AI tool
an unnamed/unconfirmed AI tool. Note: Charlotin's public database records tool attribution only where a court order, brief, or reporting on the matter states it explicitly; "unidentified" or "implied" means the record indicates AI use but does not name a specific product — we do not guess.
Outcome
Not specified in source record.
Additional detail
The district court found multiple case citations in Plaintiff’s counsel’s Opposition could not be located on Westlaw and appeared to be fake (suggesting improper use of AI). The court explained that citation to non-existent opinions undermines filings and may violate Rule 11, but because it was transferring the case to Arizona it left investigation or sanctioning to the receiving judge rather than imposing relief itself.
How Citation Safe would have caught this
Citation Safe runs three deterministic layers before a brief is filed: (1) does the citation exist against CourtListener's database of published opinions, (2) if quoted, does that exact language appear in the source, (3) does the cited case actually support the proposition it is cited for. Fabricated case citations fail Layer 1. Fabricated or misattributed quotations fail Layer 2 even when the underlying case is real. Misrepresented holdings — a real case cited for a proposition it does not support — are the target of Layer 3. None of these checks involve asking another language model whether the citation looks right; they are lookups and text-matches against the actual source, which is why a hallucinated citation has to survive a direct lookup against the authoritative source — not another model's opinion — to earn a VERIFIED stamp; our measured false-verify rate is published live at /quality.
Check a brief before you file it → · See our live false-verify rate
Source: https://www.damiencharlotin.com/documents/2742/Carey_v._Breakell_et_al_USA_30_July_2026.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).