Skip to content
Submit incident
AII-1001-96563
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.Supporting sources are pending verification.

Utah lawyer sanctioned for ChatGPT-generated false legal citations

The Utah court of appeals sanctioned lawyer Richard Bednar after discovering his petition for appeal contained false legal citations, including a nonexistent case, that were generated by ChatGPT. Bednar acknowledged the errors and stated an unlicensed law clerk had prepared the brief without his independent verification of its accuracy.

April 2025 Β· Richard Bednar Β· Reliability and hallucination
April 2025
AII-1001-96563
Quick facts
Category
Reliability and hallucination
Country
United States
Date
April 2025
Drafted by
Raidu Incident Detection

What happened

An unlicensed law clerk using ChatGPT prepared a petition for interlocutory appeal filed by Richard Bednar and Douglas Durbano in the Utah court of appeals. The brief contained false citations, including a nonexistent case titled 'Royer v Nelson' that could only be found in ChatGPT and references to cases wholly unrelated to the subject matter.

Bednar acknowledged the errors and apologized, stating that he did not independently check the accuracy of the petition before filing. The law clerk was terminated from the law firm.

Reported impact

The Utah court of appeals sanctioned Bednar for filing the brief with fabricated legal authority obtained from ChatGPT. Bednar offered to pay related attorney fees to make amends.

Index analysis

The court emphasized that while AI is a legal research tool that will continue to evolve, every attorney remains responsible for the accuracy of their filings.

Classification

Category
Reliability and hallucination
Industry
Government, security and defence
Country
United States
Harm status
Actual harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Reputational
Failure modes
Fabricated citation, Hallucination, Overreliance
AI systems
ChatGPT

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident