Federal judges admit AI tools produced erroneous court filings with fake citations
Two federal judges in New Jersey and Mississippi admitted their offices used AI tools that generated fabricated case citations and fake party references in court orders. The erroneous documents were docketed and later retracted after defendants flagged the errors, prompting scrutiny from the Senate Judiciary Committee on judicial oversight of AI use.
What happened
Judge Julien Xavier Neals in New Jersey admitted that a law school intern used ChatGPT to perform legal research, resulting in a June 30 order containing case quotations that did not exist.
Judge Henry T. Wingate in Mississippi admitted that a law clerk used Perplexity as a drafting assistant, resulting in a July 20 temporary restraining order that referred to parties, allegations, and quotes unconnected to the case. Wingate wrote that the draft 'should have never been docketed' and attributed the error to 'a failure to put the draft opinion through the final review process.'
Reported impact
Both faulty court documents were docketed and had to be hastily retracted after defendants alerted the judges to the errors.
The mistakes raised concerns about judicial accountability and oversight of AI use in federal courts, with Senate Judiciary Committee Chairman Chuck Grassley stating he had 'never seen or heard of anything like this from any federal court.'
Index analysis
The AI tools generated hallucinated case citations and fabricated references to parties and allegations that were not part of the actual cases being decided.
Neither judge explained the cause of the errors until the Senate Judiciary Committee contacted them, raising questions about how much judges scrutinize documents issued under their names.
Governance controls
Neals said he established a written policy in his chambers prohibiting the use of generative AI in legal research or drafting court filings.
Both judges said they took corrective measures and will implement additional reviews of court filings before they are submitted. The Administrative Office of the U.S. Courts issued interim guidance in July suggesting users 'consider whether the use of AI should be disclosed' in judicial functions and established a task force to issue additional guidance on AI use in federal courts.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.