Sanction teardown · E.D. Arkansas, USA · 2025-03-03
Nguyen v. Savage Enterprises
What happened
In E.D. Arkansas, 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:
-
Fabricated (Case Law)Plaintiff cited nonexistent authority in support of her response to the motion to dismiss, leading to Rule 11(b) sanctions.
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
Monetary sanction (monetary penalty: 1000 USD.)
Additional detail
AI UseNguyen did not confirm which AI tool was used but acknowledged that AI “may have contributed.” The court inferred the use of generative AI from the pattern of hallucinated citations and accepted Nguyen’s candid acknowledgment of error, though this did not excuse the Rule 11 violation.Hallucination DetailsFictitious citations included:Kraft v. Brown & Williamson Tobacco Corp., 668 F. Supp. 2d 806 (E.D. Ark. 2009)Young v. Johnson & Johnson, 983 F. Supp. 2d 747 (E.D. Ark. 2013)Carpenter v. Auto-West Inc., 553 S.W.3d 480 (Ark. 2018)Miller v. Hall, 360 S.W.2d 704 (Ark. 1962)None of these cases existed in Westlaw or Lexis, and the quotes attributed to them were fabricated.Outcome / SanctionThe court imposed a $1,000 monetary sanction on Counsel for citing non-existent case law in violation of Rule 11(b). It found her conduct unjustified, despite her apology and explanation that AI may have been involved. The court emphasized that citing fake legal authorities is an abuse of the adversary system and warrants sanctions.
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/341/Nguyen_v._Wheeler_E.D._Arkansas_USA_March_3_2025.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).