ChatGPT lands NYC-based lawyer in hot water for 'fake legal citations'

Schwartz filed a 10-page legal brief in which he cited multiple court decisions as citations, all of which turned out to be fake. (Image: Gabby Jones/Bloomberg)

ChatGPT has landed a New York City-based lawyer in hot water after it fabricated a series of legal cases that he used as citations in his filing.

Steven Schwartz from the law firm Levidow and Oberman used ChatGPT for assistance in writing a legal brief.

Also Read | ChatGPT: Now available in iOS in India, supports Apple's voice dictation

As reported by Engadget, the firm was suing Columbian airline Avianca on behalf of its client, Roberto Mata, who suffered an injury sustained from a serving cart while on the airline in 2019, claiming negligence by an employee. Schwartz filed a 10-page legal brief in which he cited multiple court decisions as citations, all of which turned out to be fake.

As you might imagine, Judge Kevin Castel of the Southern District of New York was less than impressed.

Schwartz submitted an affidavit claiming he used ChatGPT to supplement research for his case, and said that he was, "unaware of the possibility," that the bot's content, "could be false".

He also shared screenshots in which he asked ChatGPT to verify the credibility of the sources in his citations and the bot responded by saying the cases were real. They weren't.

Also Read | ChatGPT of investment advice is here, JPMorgan trademarks 'IndexGPT'

This embarrassing goof up has now put Schwartz's legal career in jeopardy and he now faces sanctions for submitting falsified information to the court.

Schwartz said he, "greatly regrets" using ChatGPT and has vowed to never use it again, "without absolute verification of its authenticity".

For now, the judge has ordered another hearing to discuss sanctions for the "unprecedented circumstance".