The relevant events began on March 1, when the plaintiff’s lawyers submitted an “affirmation” in opposition to the defendant’s motion to dismiss, which first cited the fake cases. On March 15, the defendant a reply brief questioning the existence of many of the cases cited in plaintiff’s “affirmation.”
After reading the reply and being unable to locate a number of the cases himself, Judge Castel ordered plaintiff’s lawyers to produce the cases cited in their opposition. Plaintiff’s counsel submitted an affidavit attaching the excerpts of the “cases” on April 25th. Judge Castel reviewed the “purported decisions” and described them as showing “stylistic and reasoning flaws that do not generally appear in decisions” from federal courts and containing legal analysis that was “gibberish.” On May 4, he ordered plaintiff’s lawyers to show cause why they should not be sanctioned. On May 25, one of the lawyers finally admitted that he had used ChatGPT to conduct his legal research, not understanding that it could invent fake cases.