Michael Cohen, the former lawyer for Donald Trump, admitted to citing fake, AI-generated court cases in a legal document that wound up in front of a federal judge, as reported earlier by The New York Times. A filing unsealed on Friday says Cohen used Google’s Bard to perform research after mistaking it for “a super-charged search engine” rather than an AI chatbot.

The document in question was a motion that asked a federal judge to shorten the length of Cohen’s three-year probation, which he’s now facing following prison time and a guilty plea to tax evasion and other charges. But after reviewing the letter brief, US District Judge Jesse Furman wrote in a filing that “none of these cases exist” and asked Cohen’s lawyer, David Schwartz, to explain why the three cases are included in the motion as well as whether his now-disbarred client helped draft it.

In response, Cohen submitted a written statement saying he didn’t intend to mislead the court, adding that he used Google Bard to do legal research and sent some of his findings to Schwartz. However, Cohen says he didn’t realize the cases cited by Bard had the potential to be fake, nor did he think Schwartz would add the citations to the motion “without even confirming that they existed.” Schwartz facing potential sanctions for including the phony citations.

“As a non-lawyer I have not kept up with emerging trends (and related risks) in legal technology and did not know that Google Bard was a generative text service that, like Chat-GPT, could show citations and descriptions that looked real but actually were not,” Cohen writes. “Instead, I understood it to be a super-charged search engine and had repeatedly used it in other contexts to (successfully) find accurate information online.”