9-6-2023 (NEW YORK) A New York-based lawyer, Steven Schwartz, found himself in an embarrassing situation this week when he had to apologize to a judge for submitting a court filing filled with falsehoods generated by the AI chatbot, ChatGPT. The mishap occurred during a civil case in Manhattan federal court involving a man named Roberto Mata, who is suing Colombian airline Avianca for injuries sustained during a flight from El Salvador to New York in August 2019.
After Avianca’s legal team requested the dismissal of the case, Schwartz filed a response citing several supposed legal decisions to support his argument in favor of proceeding with the litigation. Among the cases mentioned were Petersen v. Iran Air, Varghese v. China Southern Airlines, and Shaboon v. Egyptair. The Varghese case even included specific dates, internal citations, and quotes.
However, a significant problem emerged: neither Avianca’s lawyers nor the presiding judge, P. Kevin Castel, could locate any of the cited cases. It turned out that ChatGPT had invented the entire legal framework.
In response to this revelation, Judge Castel expressed astonishment, stating, “The court is presented with an unprecedented circumstance.” He observed that six of the submitted cases appeared to be fictional, complete with fabricated quotes and citations.
Consequently, the judge ordered Schwartz and his law partner to appear before him to potentially face sanctions for their actions.
Ahead of the hearing, in a filing submitted on Tuesday (Jun 9), Schwartz expressed deep regret and issued a sincere apology to the court for his “deeply regrettable mistake.” He explained that his college-educated children had introduced him to ChatGPT, and it was the first time he had utilized it for his professional work.
Schwartz further admitted that he had believed ChatGPT to be a reliable search engine at the time of conducting legal research for the case. He acknowledged his incorrect assumption and emphasized that his intention had never been to mislead the court.
Since its launch last year, ChatGPT has gained worldwide attention for its ability to produce human-like content, ranging from essays to poems and conversations, all generated from simple prompts. This technology has led to an increase in generative AI content, leaving lawmakers grappling with how to effectively regulate such bots.
At the time of reporting, OpenAI, the organization behind ChatGPT, had not responded to requests for comments on Schwartz’s predicament.
As news of the incident broke, Schwartz and his law firm, Levidow, Levidow & Oberman, found themselves subjected to public ridicule in media coverage. Schwartz expressed his deep embarrassment on both a personal and professional level, as he realized that these articles would remain accessible for years to come.
Assuring the court that he had learned a valuable lesson, Schwartz pledged never to commit a similar error again.
