Lawyers Tried ChatGPT In A Real Case - It Backfired

ChatGPT, developed by OpenAI, has astounded users with its versatile capabilities, ranging from writing essays and speeches to assisting with homework, coding, and even composing poems. However, despite its impressive feats, it has shown a tendency to fabricate information.

Recently, two lawyers from New York faced penalties for their use of ChatGPT in a legal case. Steven Schwartz, a lawyer at Levidow, Levidow & Oberman, acknowledged employing the AI chatbot to conduct research for a personal injury lawsuit against Avianca, an airline. The intention was to locate legal precedents that could support their client's case. Unfortunately, during court proceedings, the lawyers representing Avianca were unable to verify several case citations provided by Schwartz, as they were found to be mostly fictitious. Some of these references were entirely fabricated, while others misidentified judges or involved airlines that didn't even exist.

District judge Peter Kevin Castel condemned Schwartz and his colleague, Peter LoDuca, whose name appeared on the brief, for acting in bad faith. The judge accused them of deliberately evading the truth and presenting misleading statements to the court. He further criticized portions of the submitted brief as "gibberish" and "nonsensical," even including fake quotes.

Steven Schwartz, of law firm Levidow, Levidow & Oberman. Photo by Steven Hirsch

One of the challenges with generative AI, like ChatGPT and Google's Bard, is its tendency to "hallucinate" responses, as it may lack a genuine comprehension of the data it has been fed. This raises concerns among those worried about the potential spread of disinformation.

When asked about its appropriateness for writing a legal brief, ChatGPT itself responded by stating that while it can offer general information and assistance, it is crucial to recognize that it is merely an AI language model and not a qualified legal professional.

Judge Castel expressed concern about lawyers using AI for assistance, deeming it "inherently improper." He emphasized their responsibility to ensure the accuracy of their filings. The lawyers in this case persisted in defending the fake opinions even after both the court and the airline raised questions.

As a consequence of their actions, Schwartz, LoDuca, and their law firm were collectively fined $5,000 (£3,926). Levidow, Levidow & Oberman is contemplating an appeal, stating that they made a sincere mistake by not considering that a technological tool could fabricate legal cases entirely.