You said:
Can a Utah lawyer avoid sanctions for fabricated citations by explaining that a law clerk used ChatGPT without the lawyer's knowledge?
ChatGPT said:

No. A Utah lawyer cannot automatically avoid sanctions by claiming that a law clerk used ChatGPT without the lawyer's knowledge. The explanation may affect the severity of sanctions, but it does not eliminate the lawyer's independent responsibility to verify citations submitted to a court.

Three authorities are particularly relevant.

1. Utah Rule of Civil Procedure 11

Under Utah R. Civ. P. 11(b), an attorney presenting a pleading, motion, or other paper to the court certifies that, after an inquiry reasonable under the circumstances:

The legal arguments are warranted by existing law or a nonfrivolous argument for changing it.

The factual contentions have evidentiary support.

The certification applies when the attorney signs, files, submits, or later advocates the document.

Crucially, Rule 11(c) permits sanctions against an attorney responsible for a violation. Delegating research to a law clerk does not transfer the attorney's certification obligations to that clerk.

However, Rule 11 is not a strict-liability rule. The court must consider whether the attorney's inquiry was objectively reasonable under the circumstances. A lawyer's lack of knowledge may therefore matter, particularly if reasonable verification procedures were followed.

2. Utah Rules of Professional Conduct

Two rules are especially important:

Rule 5.3: Responsibilities Regarding Nonlawyer Assistance

Lawyers must make reasonable efforts to ensure that nonlawyer assistants' conduct is compatible with the lawyer's professional obligations.

If the clerk was a nonlawyer employee, the supervising attorney's responsibilities include appropriate oversight. If the clerk was a licensed lawyer, Rule 5.1 concerning supervision of subordinate lawyers may apply instead.

Rule 3.3: Candor Toward the Tribunal

A lawyer must not knowingly make false statements of fact or law to a tribunal and must correct certain false statements previously made.

Importantly, Rule 3.3's knowledge requirement differs from Rule 11's objective reasonableness standard. A lawyer might avoid a finding of knowing misconduct while still facing Rule 11 sanctions for failing to verify citations.

3. Federal precedent: Mata v. Avianca

In Mata v. Avianca, Inc., 678 F. Supp. 3d 443 (S.D.N.Y. 2023), attorneys were sanctioned after submitting fabricated judicial decisions generated by ChatGPT.

The court emphasized not merely the initial use of AI, but the attorneys' failure to verify authorities and their subsequent conduct when the authenticity of those authorities was questioned.

This is a federal district court decision from New York, not binding Utah precedent, but it illustrates why blaming an AI system or the person who operated it does not necessarily excuse the attorney who submitted the document.

The decisive distinction

Circumstance

	

Likely significance




Clerk secretly used ChatGPT

	

May mitigate culpability




Attorney reasonably verified every citation

	

Strong argument against sanctions




Attorney signed without checking authorities

	

Significant sanctions exposure




Attorney learned citations were false and promptly corrected them

	

Potential mitigation




Attorney defended fabricated citations after receiving notice

	

Substantially greater exposure

Bottom line

A Utah lawyer's strongest defense would not be "I didn't know my clerk used ChatGPT."

It would be "I conducted a reasonable inquiry before submitting the document, and here is the evidence of that inquiry."

The distinction matters because Rule 11 evaluates the reasonableness of the attorney's conduct, not simply whether the attorney personally intended to mislead the court.

One qualification: Utah state-court Rule 11, federal Rule 11, appellate briefing rules, and a court's inherent sanctioning authority can impose different requirements. The procedural setting matters.

I cannot verify Utah decisions issued through November 4, 2026, because live web search is unavailable in this conversation. Accordingly, this is an analysis of the governing rules and established federal authority, not a claim that no newer Utah decision addresses this precise defense.

ChatGPT can make mistakes. Check important info.
Latest response




Thinking effort
Medium