AI news ·
Federal judge sanctions attorney for citing fictitious AI-generated case in TCPA lawsuit
A Texas judge sanctioned a lawyer for citing a fake AI case, adding to 434 tracked hallucination incidents. She must now verify all citations and complete ethics courses.
A federal judge in Texas sanctioned a TCPA plaintiff's attorney for submitting a brief that cited a non-existent case produced by generative AI, issuing financial penalties and a public reprimand in an order dated June 11, 2026. U.S. District Judge Amos L. Mazzant found that Amy L.B. Ginsburg violated Federal Rule of Civil Procedure 11 in the case Matt McCormick v. Texakoma Financial, Inc., marking the first time generative AI sanctions have hit a TCPA lawsuit while the broader tally of such incidents accelerates.
A database tracking AI hallucination cases among U.S. lawyers nearly doubled from 239 in February 2026 to 434 in June 2026, according to data the Fifth Circuit cited and which is compiled in the AI Hallucination Cases database maintained by Damien Charlotin.
The fictitious case and shifting blame
Defendant Texakoma Financial moved for summary judgment after plaintiff Matt McCormick claimed three calls violated the TCPA. McCormick's response brief included a fictitious case name, fabricated quotes, and unsupported legal propositions. When the defendant flagged the problems, Ginsburg filed no reply. The court then ordered her to appear in person with copies of the cases she had cited.
In a response filed within an hour of the hearing, Ginsburg accepted responsibility but blamed her law clerk and a paralegal. She said the clerk drafted the brief using AI, that she recognized a case name from her husband's work and corrected it, and that a paralegal mistakenly filed an older draft containing the hallucinated citation. At the hearing, however, Ginsburg said she could not confirm AI was used because the clerk had left her firm. She provided no evidence that a corrected final draft existed and offered only highlighted copies of the case she assumed the clerk meant to reference.
Court finds explanations "lack credibility"
Judge Mazzant rejected the blame-shifting. He wrote that someone who cites a nonexistent case "the case name, number, year, and quotation… [is] not struggling with research or citation formatting. They are struggling with honesty." The court also found it suspect that Ginsburg described the clerk having trouble locating the "correct case or cite," treating two different errors as interchangeable. "These are not interchangeable details," the order said, and discovering a wholly fictitious case "would warrant a substantially different reaction than discovering, say, a typo in the Westlaw reporter number."
The judge noted that other attorneys have tried similar excuses, including blaming paralegals for filing wrong documents. Ginsburg claimed she was unaware of such patterns despite practicing for more than 20 years and attending "numerous CLEs on the use of AI." The court concluded her story was "a contrived attempt to shift the blame to others" and that her attempts to take ownership "at best, lack credibility."
Sanctions designed to deter
The court imposed a series of measures beyond financial sanctions. Ginsburg must verify the existence and accuracy of every legal authority in all filings this year. If any fictional authority or material misrepresentation appears, she has three days to notify the court and opposing parties to correct the record. She must also complete CLE courses on ethical AI use and include a signed Certification of Verification in every filing that cites legal authority. That certification must state:
"I, Amy L.B. Ginsburg, certify that I personally reviewed each citation and authority referenced in this filing and that, to the best of my knowledge and belief, all citations are genuine and accurately reflect the law as of the date of filing. I also certify that I personally verified that the file I reviewed is the same one that is being submitted to this Court."
The order will be published in the Federal Supplement and distributed to all district and magistrate judges in the Eastern District of Texas.
Why this matters for legal professionals
This case shows that relying on generative AI without rigorous, personal verification can lead to sanctions, public reprimand, and ongoing court oversight. Attorneys must confirm every citation, ensure the filed document matches the reviewed draft, and scrutinize opposing filings for fabricated authorities. Understanding how large language models generate text is now a core competency-AI for Legal Professionals Courses cover ethical use and verification workflows, while Generative AI and LLM Courses explain how hallucinations occur and how to detect them. When errors happen, shifting blame to junior staff or technology only compounds the damage. The court's reaction makes clear that genuine accountability, not a string of excuses, is what separates an honest mistake from sanctionable conduct.