Judge Disqualifies All Lawyers After Both Sides Used AI

A federal judge in Mississippi has taken the extraordinary step of disqualifying every lawyer involved in a case after discovering that attorneys on both sides used artificial intelligence tools that generated fictitious legal citations.

Senior U.S. District Judge Sharion Aycock canceled the trial in a contractual dispute between lawyer Tom Withers and the city of Aberdeen, Mississippi, over unpaid legal fees. She sanctioned all four attorneys involved, fining them between $1,000 and $3,500 each and barring two from appearing before her court for two years.

When AI argues against itself

The case represents what one observing attorney called a "comedy of AI errors"—a situation where generative AI tools were effectively used to argue against each other. Both sides submitted legal briefs containing hallucinated case citations, nonexistent precedents that AI language models fabricated.

At a January hearing, all four lawyers either admitted to directly using AI for legal research or acknowledged they had rubber-stamped AI-generated briefs without proper review. Attorney Kathleen Wilson told the court she used an AI tool called First Drafts to write an entire briefing. Another attorney, Kathryn Williams, admitted using an unnamed in-house legal research tool that wasn't designed to cover Mississippi law—despite this being a Mississippi case.

Continued violations after detection

Judge Aycock expressed particular concern about Wilson's conduct. After being caught and questioned about AI-generated hallucinations in January, Wilson claimed she didn't know AI could produce fictitious cases. The judge found this explanation "insufficient and incredulous."

More troubling, other judges detected hallucinated cases in Wilson's filings as recently as April—four months after she was initially confronted. "Her continued AI misuse demonstrates an extreme dereliction of professional responsibility," Aycock wrote, adding that Wilson's apologies "were not sincere."

The judge noted that Williams used a legal research tool knowing it wasn't built to produce results for Mississippi jurisdictions, yet she deployed it anyway in what she described as her only Mississippi case.

Why it matters

This case marks a significant escalation in judicial response to AI misuse in legal practice. Rather than sanctioning individual attorneys, the judge removed all counsel from both sides—effectively resetting the entire case. The decision signals that courts are moving beyond warnings to concrete consequences, including multi-year bans from practice. For law firms and corporate legal departments, the ruling underscores that AI tools require rigorous verification protocols, and that claiming ignorance about AI limitations is no longer a viable defense. The fact that one attorney continued using AI improperly after being caught suggests the legal profession still lacks adequate safeguards and training around these technologies.

A pattern of judicial frustration

The Mississippi case follows a growing pattern of judges confronting AI hallucinations in court filings nationwide. Aycock wrote that her court was "yet again burdened with addressing AI hallucinations in court filings," describing the situation as "rampant unverified AI usage within the legal field."

In her sanctions order, she characterized the case as "a prime example of the risk associated with serving as a rubber-stamp" for AI-generated content.

The details were first reported by 404 Media, which has documented multiple instances of lawyers submitting AI-generated filings containing fabricated case law.