ChatGPT AI Usage for Case Research Leads to Fake Court Citations, US Lawyer Admits

A US lawyer admitted to using ChatGPT AI for case research, leading to the inclusion of fake court citations.

The disclosure comes to light amid the ongoing Avianca Airlines lawsuit, wherein multiple erroneous findings by the viral AI chatbot made it to court.

ChatGPT Lawyer Case Research

According to The Verge, a United States lawyer Steven A. Schwartz has come clean in his recent affidavit, in which he confessed that he used the popular OpenAI chatbot, ChatGPT, for his case research.

CNN reports that the Levidow, Levidow, & Oberman attorney represents Roberto Mata, who brought Avianca airline to court over injuries he allegedly sustained while serving cart on the airline in 2019.

Schwartz, a lawyer for more than three decades, ended up citing a couple of fake court citations, United States District Judge Kevin Castel confirmed. He says, "Six of the submitted cases appear to be bogus judicial decisions with bogus quotes and bogus internal citations."

The US lawyer tried to verify if the court citations were real. His affidavit shows that he asked the AI-powered tool if it was a "real case," to which the chatbot's response was a resounding yes. But he seems unconvinced at first. So he proceeded to ask ChatGPT for its source. Interestingly, the chatbot cited Westlaw and LexisNexis for the case citations.

With the bogus source, he seemed satisfied. Yet, he still asked the AI chatbot if the other cases cited were fake, and ChatGPT told him they were authentic.

Lawyer Apologizes for Fake AI-Generated Court Citations

Schwartz has since apologized for using an AI chatbot for his recent court research, which cited half a dozen "bogus" court decisions.

Straits Times reports that the New York lawyer, Schwarz, told the Judge that his intentions were clean, noting that he did not deliberately try to deceive the airline or the court. He also disclosed that it was the first instance he used the viral AI-powered chatbot. The attorney says he "was unaware of the possibility that its content could be false."

And as such, the New York attorney has apologized for depending on ChatGPT, saying that he "greatly regrets" his actions. He also assures that he will only use the AI chatbot again after verifying the integrity of its responses.

CNN notes that the fellow attorney of Schwartz, in the case, Peter Doduca, says he "had no reason to doubt the sincerity." And on top of that, he cleared out that he was not part of the case research, which included the bogus case citations.

Tags

AI