Sanction teardown · CA Louisiana, USA · 2025-10-23

In re: Sanctions Order of Kenney

What happened

In CA Louisiana, USA, a filing relied on ChatGPT, Microsoft Copilot, Google to help draft legal argument. The court identified the following problems with the citations in that filing:

• Fabricated (Case Law)
  Court found 'Burns v. State, 813 So.2d 574 (La. App. 1 Cir. 2001)' was entirely fabricated by generative AI and did not exist; reporter/page numbers pointed to unrelated cases.
• Fabricated (Case Law)
  Court found 'Smith v. Christus St. Patrick Hosp., 20-306 (La. App. 3 Cir. 11/3/20), 308 So.3d 1216' was fabricated by AI and the reporter/page numbers referenced an unrelated case.
• Fabricated (Case Law)
  Court found 'Doe v. Southwest Louisiana Hosp. Ass'n, 02-180 (La. App. 3 Cir. 7/10/02), 833 So.2d 1136' was fabricated by AI; referenced reporter/page numbers pointed to unrelated decisions.
• False Quotes (Case Law)
  Court determined plaintiff's filing contained a false/altered quotation attributed to Thomas v. Reg'l Health Sys. of Acadiana and an incorrect reporter citation; quotation not found in the actual opinion.
• Misrepresented (Case Law)
  Court found counsel misrepresented the holding of Acara v. Banks by asserting it supports that HIPAA violations may serve as evidence of standard of care in state tort claims; Acara does not support that proposition.

Which AI tool

ChatGPT, Microsoft Copilot, Google. 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

Costs order, 3 hours CLE on ethical use of generative AI, referral to Office of Disciplinary Counsel (monetary penalty: 1368 USD.)

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/899/In_re_Sanctions_Order_of_Kenney_USA_23_October_2025.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).