Federal judge in Mississippi admits staff used AI to draft inaccurate order
ftwitty via Getty Images
‘Given that I hold myself and my staff to the highest standards of conduct, I do not expect that a mistake like this one will occur in the future.’ – Judge Henry T. Wingate, in response to an inquiry by U.S. Sen. Chuck Grassley
This story was originally published by Mississippi Today.
A federal judge in Mississippi has admitted that his staff used artificial intelligence to draft a flawed court order, after months of speculation and an inquiry from a U.S. senator.
U.S. District Judge Henry T. Wingate sent a letter on Tuesday to the director of the Administrative Office of the Courts in response to an inquiry from Senate Judiciary Committee Chairman Chuck Grassley. Wingate acknowledged that his law clerk used an AI program to compose an order that paused the enforcement of a state law that bans diversity, equity and inclusion programs in Mississippi public schools.
The order, issued July 20, was factually inaccurate — naming defendants and plaintiffs that weren’t parties to the case, misquoting state law and referencing a case that doesn’t exist — which led the Mississippi Attorney General’s Office to raise concerns.
Wingate then replaced the order with a corrected version, wiped the flawed order from the docket, and denied a request from the Attorney General’s Office to restore the original order with errors to the public docket. He refused to explain the errors, calling them “clerical” mistakes.
Months later, Grassley sent a letter to Wingate, asking him to address the errors and the speculated use of AI.
Wingate finally acknowledged that one of his law clerks used an AI program called Perplexity to analyze publicly available information from the docket to craft the order. But Wingate emphasized no privileged or sealed information was used. Wingate says he still does not plan to restore the original order to the docket, in an effort to avoid confusion.
The judge conceded that the order was a draft opinion, and it did not undergo proper review before being added to the docket. Wingate said he’s taking steps to make sure this doesn’t happen again.
The corrective measures include requiring all draft opinions, orders and memos to undergo a second, independent review. He will also require all cited cases be printed and attached to the final draft.
“I manage a very busy docket and strive to maintain the public’s trust by administering justice in a fair and transparent manner,” Wingate wrote. “Given that I hold myself and my staff to the highest standards of conduct, I do not expect that a mistake like this one will occur in the future.”
Grassley, a Republican from Iowa, commended Wingate, a Reagan-appointed judge, for acknowledging the mistake.
“Each federal judge, and the judiciary as an institution, has an obligation to ensure the use of generative AI does not violate litigants’ rights or prevent fair treatment under the law,” Grassley said in a statement. “The judicial branch needs to develop more decisive, meaningful and permanent AI policies and guidelines.”
Still, AI usage in the federal judiciary is a serious issue, and it’s largely uncharted territory, especially among federal judges.
The legal profession has seen a rise in AI use in recent years, with people relying on software or processes that attempt to replicate aspects of human work. These use vast amounts of data to accomplish tasks such as researching court cases and citing them in legal briefs.
But these systems are not perfect and can “hallucinate,” or provide false information.
Judges in Mississippi have punished attorneys in both federal and state courts for using AI because attorneys have an ethical and professional responsibility to tell the truth in court.
But there’s little accountability when the roles are reversed.
Robert Conrad Jr., the director of the federal Administrative Office of the Courts, also sent Grassley a letter noting that he recently established an AI task force composed of judges and technology experts to recommend new policies on AI usage in the federal court system.
Conrad said the task force, during the summer, published interim guidance on AI, suggesting that attorneys should review and independently verify all AI-generated content and should consider disclosing when they’ve used AI to help craft documents.
This article first appeared on Mississippi Today and is republished here under a Creative Commons Attribution-NoDerivatives 4.0 International License.
