Jerome Dewald sat down with her legs crossed and hands beat in their womb before a jury of appeal of judges of the State of New York, ready to support the reversal of the decision of a lower court in its dispute with an ex -employer.
The Court had allowed Mr. Dewald, who is not a lawyer and was representing, to accompany his argument with a pre -aggravated video presentation.
When the video began to play, he showed an apparently younger man than the 74 years of Mr. Dewald who wore a blue shirt and a beige sweater and standing in front of what seemed to be a blurred virtual background.
A few seconds in the video, one of the judges, confused by the image on the screen, asked Mr. Dewald if the man were his lawyer.
“I generated it,” Dewald replied. “This is not a real person.”
The judge, the judge Sallie Manzanet-Daniels of the first judicial department of the appeal division, stopped for a moment. It was clear that he was a discontent of his answer.
“It would have been nice to know that when you asked your question,” he took against him.
“I don’t appreciate that I was misleading,” he added before screaming for someone who turns the video off.
What Mr. Dewald did not reveal was that he had created the digital avatar using artificial intelligence software, the last example of AI which insinuates itself into the American legal system in potentially worrying ways.
The hearing in which Mr. Dewald made his presentation, on March 26, was filmed by cameras of the judicial system and previously reported by the Associated Press.
Friday reached, Mr. Dewald, the actor in case, said he had been overwhelmed by embarrassment in the hearing. He said he had sent the judges a letter of apologies shortly after, expressing his deep regret and recognizing that his actions had “inadvertently missed” the court.
He said he had resorted to the use of the software after having stumbled on his words in previous legal proceedings. Using artificial intelligence for the presentation, he thought, he could alleviate the pressure he felt in the classroom.
He said he had planned to make a digital version of himself, but had encountered “technical difficulties” in doing it, which prompted him to create a false person for recording.
“My intent was never to deceive, but rather of presenting my arguments in the most efficient way possible,” he said in his letter to the judges. “However, I recognize that the correct dissemination and transparency must always have precedence.”
Autodescribed entrepreneur, Dewald was appealing to a previous sentence in a contractual dispute with an ex -employer. In the end he presented an oral topic at the appeal hearing, stammering and making frequent breaks to group and read observations prepared by his mobile phone.
As embarrassed, Mr. Dewald could get comforting that the real lawyers have put themselves in trouble for using the IA in court.
In 2023, a New York lawyer faced serious repercussions after using chatgpt to create a short patent full of false judicial opinions and legal quotations. The case showed the defects in relying on artificial intelligence and reverberated during legal trade.
The same year, Michael Cohen, a former lawyer and fixator for President Trump, provided his lawyer to the false legal quotes he had obtained from Google Bard, an artificial intelligence program. Mr. Cohen asked for pity from the federal judge who presides over his case, underlining that he had not known that the generative text service could provide false information.
Some experts say that artificial intelligence and large models can be useful to people who have legal issues to face but cannot afford lawyers. However, the risks of technology remain.
“They can still hallucinated – producing very convincing -aspect information” that are actually “false or meaningless”, said Daniel Shin, assistant to the director of research at the Center for Legal and Court Technology at William & Mary Law School. “This risk must be addressed.”