Senior U.S. District Judge P. Kevin Castel, presiding over the Southern District of New York, ordered two lawyers and their law firm to provide a compelling justification for why they should not be sanctioned. The lawyers allegedly submitted a brief containing citations to fake cases, as revealed through research conducted by ChatGPT, an advanced language model developed by OpenAI.
The judge’s decision resulted from an order issued on May 4, wherein Judge Castel highlighted that the law firm’s filing was “replete with citations to nonexistent cases.” The judge further instructed one of the lawyers to submit an affidavit containing the referenced opinions. However, upon examination, it was discovered that six of the cited decisions “appear to be bogus,” complete with “bogus quotes and bogus internal citations,” as stated by Judge Castel.
According to an affidavit filed on May 25 by attorney Steven A. Schwartz from Levidow, Levidow & Oberman, it was revealed that the fake cases were provided by ChatGPT. Schwartz, who has been practicing law in New York for over three decades, admitted that he had not previously utilized ChatGPT as a source for legal research and was unaware of the possibility of its content being false.
Schwartz explained in his affidavit that ChatGPT had assured him that the cited cases were real and could be found in reputable legal databases such as LexisNexis and Westlaw. He had relied on these assurances while submitting the brief. Another attorney who co-signed the brief, Peter LoDuca, reportedly had no knowledge of Schwartz’s research method. LoDuca became the attorney of record after their case was transferred to the Southern District of New York, where Schwartz had not obtained admission.
To address this concerning situation, Judge Castel has scheduled a show cause hearing for June 8, as stated in an order issued on May 26. The case has gained attention from various publications, including the New York Times and the Volokh Conspiracy, which have provided coverage on the matter. The Volokh Conspiracy even shared a link to a case page from CourtListener related to this case.
In response to the request for comment, Schwartz did not immediately provide a response to the ABA Journal, which reached out via email and voicemail. On the other hand, LoDuca informed the ABA Journal that he has no comment at this time.
The lawyers, Schwartz and LoDuca, represent Roberto Mata, the plaintiff in a lawsuit against Avianca Inc., an airline. Mata claims to have sustained injuries when he was struck by a metal serving cart. This real-life case involving Roberto Mata v. Avianca Inc. has attracted attention, with the New York Times noting that it exemplifies how white-collar professions still have some time before being fully replaced by robots.
Interestingly, the Volokh Conspiracy highlighted that even self-represented litigants are utilizing ChatGPT for their legal matters, indicating a broader trend of individuals turning to advanced language models for legal research.
As the case progresses, the upcoming show cause hearing on June 8 will shed more light on the lawyers’ explanation regarding the use of fake case citations. The incident serves as a reminder of lawyers’ ethical and professional responsibilities when conducting legal research, emphasizing the importance of relying on reliable and verified sources for accurate information.