Mata v. Avianca, Inc
S.D. New York · Jun 22, 2023
What happened in this matter?
Counsel filed authorities generated through ChatGPT that did not exist, then submitted purported copies after the problem was raised. The court imposed a $5,000 sanction and required notice to the client and judges whose names appeared on the fabricated opinions.
- Why the court cared
- The sanction was driven by counsel's gatekeeping failure and the decision to stand behind false material after receiving notice, not simply by the use of a generative tool.
- Why it matters now
- Mata is the operational anchor for citation verification, prompt correction, and supervising-attorney signoff. A recoverable research error became a sanctions matter because the filing and response process failed.
Why this matter is tracked
Fabricated: Case Law | Non-existent Texas appellate decision generated by ChatGPT and submitted as authority. Outcome: Monetary Fine (Lawyers & Firm); Letters to Client/Judges.
AI Use
Counsel from Levidow, Levidow & Oberman used ChatGPT for legal research to oppose a motion to dismiss a personal injury claim against Avianca airlines, citing difficulty accessing relevant federal precedent through their limited research subscription.
Hallucination Details
The attorneys' submission included at least six completely non-existent judicial decisions, complete with fabricated quotes and internal citations. Examples cited by the court include Varghese v. China Southern Airlines Co., Ltd., Shaboon v. Egyptair, Petersen v. Iran Air, Martinez v. Delta Airlines, Inc., Estate of Durden v. KLM Royal Dutch Airlines, and Miller v. United Airlines, Inc.. When challenged by opposing counsel and the court, the attorneys initially stood by the fake cases and even submitted purported copies of the opinions, which were also generated by ChatGPT and contained further bogus citations.
Ruling/Sanction
Judge P. Kevin Castel imposed a $5,000 monetary sanction jointly and severally on the two attorneys and their law firm. He also required them to send letters informing their client and each judge whose name was falsely used on the fabricated opinions about the situation.
Key Judicial Reasoning
Judge Castel found the attorneys acted in bad faith, emphasizing their "acts of conscious avoidance and false and misleading statements to the Court" after the issue was raised. The sanctions were imposed not merely for the initial error but for the failure in their gatekeeping roles and their decision to "double down" rather than promptly correcting the record. The opinion detailed the extensive harms caused by submitting fake opinions. This case is widely considered a landmark decision and is frequently cited in subsequent discussions and guidance.
When a citation is challenged, never double down — immediate candor is the only way to keep a mistake from becoming bad-faith sanctions.
Record details
What the record establishes about AI use
The attorneys admitted to using ChatGPT for research.
Opinion and Order on Sanctions issued by the U.S. District Court for the Southern District of New York following a show cause hearing.
Counsel initially defended the citations and supplied generated copies rather than promptly withdrawing or correcting them.
Tracked discrepancy record
18 citation, quotation, or authority issues are recorded in the source dataset.
- Fabricated: Case Law | Non-existent Texas appellate decision generated by ChatGPT and submitted as authority.
- Fabricated: Case Law | Non-existent Georgia appellate decision generated by ChatGPT and submitted as authority.
- Fabricated: Case Law | Internal citation in the fake Varghese opinion to a non-existent Eleventh Circuit case.
- Fabricated: Case Law | Internal citation in the fake Varghese opinion to a non-existent Fifth Circuit case; reporter cite actually corresponds to a Fourth Circuit case.
- Fabricated: Case Law | Internal citation in the fake Varghese opinion to a non-existent Eleventh Circuit case at 92 F.3d 1074.
- Fabricated: Case Law | Internal citation in the fake Varghese opinion to a non-existent Eleventh Circuit case; citation corresponds to an unrelated case.
- Misrepresented: Case Law | Cited an Eleventh Circuit Zicherman decision and citation that does not exist; real Zicherman is a Supreme Court case and unrelated Eleventh Circuit cite is Miccosukee Tribe.
- Misrepresented: Case Law | Mis-cited a bankruptcy decision and reporter; the cited Bankruptcy Reporter citation is to a different case and Second Circuit BDC 56 did not address the stay.
Show 10 additional discrepancies
- Misrepresented: Case Law | Relied on a Seventh Circuit decision that addresses Rule 11 sanctions, not the bankruptcy stay.
- Misrepresented: Case Law | Identified as a Second Circuit case and cited for the bankruptcy stay, but it is a Third Circuit decision that does not address the stay.
- Misrepresented: Case Law | Cited for a bankruptcy-stay proposition, but the case addresses whether a trustee may recover certain payments to the IRS.
- Misrepresented: Case Law | Cited for a bankruptcy-stay proposition, but the case concerned a stay pending state-court proceedings on state constitutional issues.
- Misrepresented: Case Law | Cited for a bankruptcy-stay proposition, but the case involved denial of a motion to compel arbitration.
- Fabricated: Case Law | Non-existent Eleventh Circuit opinion; Clerk confirmed no such case, docket/citation matched other matters, and the text contained gibberish analysis.
- Fabricated: Case Law | Non-existent Second Circuit decision; docket number and reporter citation belonged to unrelated cases.
- Fabricated: Case Law | Non-existent D.D.C. decision; docket and citations corresponded to different cases and even cited itself as precedent.
- Fabricated: Case Law | Non-existent Illinois appellate decision generated by ChatGPT and submitted as authority.
- False Quotes: Case Law | Quoted language about the purpose of the Montreal Convention that does not appear in the decision.
Questions this record answers
- What happened in Mata v. Avianca, Inc?
- Counsel filed authorities generated through ChatGPT that did not exist, then submitted purported copies after the problem was raised. The court imposed a $5,000 sanction and required notice to the client and judges whose names appeared on the fabricated opinions.
- Why does Mata v. Avianca, Inc matter for legal AI risk?
- Mata is the operational anchor for citation verification, prompt correction, and supervising-attorney signoff. A recoverable research error became a sanctions matter because the filing and response process failed.
- What does the public record establish about Mata v. Avianca, Inc?
- The record summarizes the outcome described in the linked public source. The recorded link is a secondary or other public source and should be checked against the docket where available. This page is not a substitute for the complete docket, subsequent history, or jurisdiction-specific advice.
- Which source supports this Mata v. Avianca, Inc summary?
- The recorded source is CourtListener / RECAP. It is classified as docket or legal-document mirror; review the linked material and subsequent docket history before relying on this summary.
Related matters
Related by court, jurisdiction, tool, or classified failure pattern. Similarity does not imply the same facts or outcome.