What was hallucinated

Fabricated: Case Law | Reply brief cited a non-existent Appellate Division decision; the Court could not locate it, ordered a copy, and counsel later admitted it was generated via ChatGPT and does not exist; Court referred counsel to the Grievance Panel.

Details

AI UseCounsel admitted using ChatGPT to find supporting case law after failing to locate precedent manually. She cited a fictitious case (Matter of Bourguignon v. Coordinated Behavioral Health Servs., Inc., 114 A.D.3d 947 (3d Dep’t 2014)) in the reply brief, never verifying its existence.Hallucination DetailsOnly one hallucinated case was cited in the reply brief: Matter of Bourguignon v. Coordinated Behavioral Health Servs., Inc., 114 A.D.3d 947 (3d Dep’t 2014). When asked to produce the case, Counsel admitted it did not exist, blaming reliance on ChatGPT.Ruling/SanctionThe Court referred Counsel to the Second Circuit’s Grievance Panel for further investigation and possible discipline. Lee was ordered to furnish a copy of the decision (translated if necessary) to her client and to file certification of compliance.Key Judicial ReasoningThe Court emphasized that attorneys must personally verify the existence and accuracy of all authorities cited. Rule 11 requires a reasonable inquiry, and no technological novelty excuses failing to meet that standard. The Second Circuit cited Mata v. Avianca approvingly, confirming that citing fake cases amounts to abusing the adversarial system.

Sanction teardown · 2nd. Cir. CA, USA · 2024-01-30

Park v. Kim

What happened

In 2nd. Cir. CA, USA, a filing relied on ChatGPT to help draft legal argument. The court identified the following problems with the citations in that filing:

Fabricated (Case Law)

Reply brief cited a non-existent Appellate Division decision; the Court could not locate it, ordered a copy, and counsel later admitted it was generated via ChatGPT and does not exist; Court referred counsel to the Grievance Panel.

Which AI tool

ChatGPT. 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

Referral to Grievance Panel + Order to Disclose Misconduct to Client.

Additional detail

AI UseCounsel admitted using ChatGPT to find supporting case law after failing to locate precedent manually. She cited a fictitious case (Matter of Bourguignon v. Coordinated Behavioral Health Servs., Inc., 114 A.D.3d 947 (3d Dep’t 2014)) in the reply brief, never verifying its existence.Hallucination DetailsOnly one hallucinated case was cited in the reply brief: Matter of Bourguignon v. Coordinated Behavioral Health Servs., Inc., 114 A.D.3d 947 (3d Dep’t 2014). When asked to produce the case, Counsel admitted it did not exist, blaming reliance on ChatGPT.Ruling/SanctionThe Court referred Counsel to the Second Circuit’s Grievance Panel for further investigation and possible discipline. Lee was ordered to furnish a copy of the decision (translated if necessary) to her client and to file certification of compliance.Key Judicial ReasoningThe Court emphasized that attorneys must personally verify the existence and accuracy of all authorities cited. Rule 11 requires a reasonable inquiry, and no technological novelty excuses failing to meet that standard. The Second Circuit cited Mata v. Avianca approvingly, confirming that citing fake cases amounts to abusing the adversarial system.

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/243/Park_v._Kim_US_2nd_Cir._CA_January_30_2024.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).