Branchoria AI Sense AI Sense

Clear, practical explanations of what AI does, where it helps, and where it fails.

Within Avianca case

What the Avianca Sanctions Changed for Lawyers

The sanctions order in Mata v. Avianca became a landmark moment in the governance of artificial intelligence because it answered a question many professionals were beginning to ask: who is responsible when AI generates false information that is later used in important work? Judge P. Kevin Castel’s answer was clear.

On this page

  • The court's finding of professional failure
  • Monetary penalties and corrective actions
  • Why the duty to verify survived AI assistance
Preview for What the Avianca Sanctions Changed for Lawyers

Introduction

The sanctions order therefore mattered far beyond one lawsuit. It established an early judicial framework for professional responsibility in the age of generative AI and became a reference point for lawyers, judges, bar associations and regulators considering how AI should be used in legal practice. [Kirkland & Ellis]kirkland.comif you think it thinks think againSource details in endnotes.

Sanctions illustration 1

What the Court Found Was the Professional Failure

Judge Castel’s opinion did not simply focus on the fact that ChatGPT produced fictional cases. The court examined the conduct of the lawyers after the errors were discovered and concluded that they had repeatedly failed in their duties to the court. According to the opinion, the lawyers filed a submission containing non-existent authorities, failed to verify those authorities through ordinary legal research methods, and continued to defend the citations after their authenticity had been questioned. [Intellectual Property Law Blog]intellectualproperty.lawsanctions handed down to lawyers who cited fake cases relying on chatgptSource details in endnotes.

A key lesson from the order is that the court viewed the problem as one of professional judgment rather than technological malfunction. Attorneys are expected to verify legal authorities before presenting them as genuine precedent. The court emphasised that established duties of competence, candour and reasonable inquiry already existed long before generative AI. The arrival of a new research tool did not eliminate those obligations. [Justia]law.justia.comMata v. Avianca, Inc., No. 1:2022cv01461 - Document 54 (S.D.N.Y. 2023):: JustiaJune 22, 2023…Published: June 22, 2023

The opinion also highlighted that the lawyers relied on ChatGPT as though it were a trustworthy legal database. Judge Castel noted that one attorney had treated the system as a kind of enhanced search engine and did not adequately investigate its limitations. The court concluded that this assumption was unreasonable given the lawyer’s responsibility to ensure the accuracy of authorities cited to a federal court. [Kirkland & Ellis]kirkland.comif you think it thinks think againSource details in endnotes.

The ChatGPT Fake Cases Sanctions Decision (Mata v. Avianca)

Channel: Lawful Masses with Leonard French

Open on YouTube

Monetary Penalties and Corrective Actions

The sanctions imposed were relatively modest in financial terms but symbolically significant. Judge Castel ordered a penalty of $5,000 against the lawyers and their law firm. More important than the monetary amount were the corrective measures designed to address the harm caused by the fabricated authorities. [CaseLaw+2Justia]caselaw.findlaw.comSource details in endnotes.

The court required the respondents to:

  • Send a copy of the sanctions order and related materials to their client.
  • Send letters to the judges whose names had been falsely attached to fabricated judicial opinions.
  • File proof that those letters had been sent.
  • Pay the monetary sanction ordered by the court. [CaseLaw]caselaw.findlaw.comSource details in endnotes.

These requirements reflected the court’s view that the damage extended beyond the immediate litigation. The fabricated opinions falsely attributed legal decisions to real judges and risked undermining confidence in the judicial system. Judge Castel warned that fake authorities could harm the reputation of courts, mislead litigants and contribute to public cynicism about legal institutions. [Justia]law.justia.comMata v. Avianca, Inc., No. 1:2022cv01461 - Document 54 (S.D.N.Y. 2023):: JustiaJune 22, 2023…Published: June 22, 2023

Sanctions illustration 2

Why the Duty to Verify Survived AI Assistance

Perhaps the most influential aspect of the decision was its treatment of AI itself. The sanctions order did not prohibit the use of generative AI in legal practice. Instead, it drew a distinction between using AI and relying on AI without verification. [Justia]law.justia.comMata v. Avianca, Inc., No. 1:2022cv01461 - Document 54 (S.D.N.Y. 2023):: JustiaJune 22, 2023…Published: June 22, 2023

The court effectively established three principles that have since shaped discussions of AI governance in professional settings:

AI can assist, but it cannot assume responsibility. The lawyer who signs a filing remains accountable for its contents. Responsibility cannot be delegated to a software system. [American Bar Association]americanbar.orggal-practice/…

Verification must be independent. One of the most striking facts in the case was that ChatGPT was reportedly asked whether the cases it generated were real, and it incorrectly confirmed that they were. The incident demonstrated that asking an AI system to validate its own output is not meaningful verification. Independent checking through authoritative sources remains necessary. [Responsible AI Labs]responsibleailabs.aiMata v. Avianca: Lawyers Sanctioned for ChatGPT-Fabricated Cases | AI Incident Watch | Responsible AI LabsJune 22, 2023…Published: June 22, 2023

Existing professional rules still apply. Judge Castel’s reasoning suggested that no special AI exception exists within lawyers’ ethical duties. New technology changes how work is performed, but not the underlying obligations of competence, diligence and honesty owed to the court. [Justia]law.justia.comMata v. Avianca, Inc., No. 1:2022cv01461 - Document 54 (S.D.N.Y. 2023):: JustiaJune 22, 2023…Published: June 22, 2023

This point proved especially important because some early discussions of AI framed hallucinations as technical errors outside normal professional responsibility. The sanctions order rejected that approach. From the court’s perspective, the central issue was not whether ChatGPT made mistakes. It was whether the lawyers fulfilled their duty to detect those mistakes before presenting them as genuine legal authority. [Justia]law.justia.comMata v. Avianca, Inc., No. 1:2022cv01461 - Document 54 (S.D.N.Y. 2023):: JustiaJune 22, 2023…Published: June 22, 2023

Practice Law With AI (Without Losing Your Law License) | Justia Webinars

Channel: Justia

Open on YouTube

Lawyer files ChatGPT DISASTER in COURT (Mata v. Avianca, Inc.)

Channel: Lawful Masses with Leonard French

Open on YouTube

How the Order Influenced AI Governance in Law

The lasting significance of Judge Castel’s order lies in the precedent it set for professional use of artificial intelligence. Courts and bar associations did not interpret the case as a reason to ban AI. Instead, it became evidence that human oversight must remain central whenever AI is used in legal work. [American Bar Association]americanbar.orggal-practice/…

The decision reinforced a governance principle that extends well beyond law: accountability follows the human decision-maker, not the software. In legal practice, that means attorneys must understand the limitations of AI tools, verify outputs through reliable sources and accept responsibility for what they submit under their names. The Mata v. Avianca sanctions order became the first widely recognised judicial statement that generative AI may change how legal research is performed, but it does not change who is responsible when that research is wrong. [Justia+2Kirkland & Ellis]law.justia.comMata v. Avianca, Inc., No. 1:2022cv01461 - Document 54 (S.D.N.Y. 2023):: JustiaJune 22, 2023…Published: June 22, 2023

Sanctions illustration 3

Amazon book picks

Further Reading

Books and field guides related to What the Avianca Sanctions Changed for Lawyers. Use these as the next step if you want deeper reading beyond the article.

Browse more on Amazon: The LegalTech Book Artificial Intelligence and Legal Analytics Law and Artificial Intelligence

As an Amazon Associate I earn from qualifying purchases.

eBay marketplace picks

Marketplace Samples

Example marketplace items related to this page. Use the search link to explore similar finds on eBay.

Using USA
Browse more on eBay.co.uk

Example items shown for inspiration; availability and pricing can change. Branchoria may earn a commission if you purchase through outbound eBay links.

Endnotes

  1. Source: law.justia.com
    Link: https://law.justia.com/cases/federal/district-courts/new-york/nysdce/1%3A2022cv01461/575368/54/
    Source snippet

    Mata v. Avianca, Inc., No. 1:2022cv01461 - Document 54 (S.D.N.Y. 2023):: JustiaJune 22, 2023...

    Published: June 22, 2023

  2. Source: kirkland.com
    Title: if you think it thinks think again
    Link: https://www.kirkland.com/publications/article/2023/07/if-you-think-it-thinks-think-again
    Source snippet

    "Kirkland & Ellis[https://www.kirkland.com/publications/article/2023/07/if-you-think-it-thinks-think-again..."](https://www.kirkland.com/publications/article/2023/07/if-you-think-it-thinks-think-again...")...

  3. Source: americanbar.org
    Link: https://www.americanbar.org/groups/young_lawyers/resources/tyl/practice-management/guide-implementing-ai-tools-legal-practice/
    Source snippet

    gal-practice/...

  4. Source: intellectualproperty.law
    Title: sanctions handed down to lawyers who cited fake cases relying on chatgpt
    Link: https://www.intellectualproperty.law/2023/06/sanctions-handed-down-to-lawyers-who-cited-fake-cases-relying-on-chatgpt/
    Source snippet

    "Intellectual Property Law Blog[https://www.intellectualproperty.law/2023/06/sanctions-handed-down-to-lawyers-who-cited-fake-cases-relying..."](https://www.intellectualproperty.law/2023/06/sanctions-handed-down-to-lawyers-who-cited-fake...

  5. Source: caselaw.findlaw.com
    Link: https://caselaw.findlaw.com/court/us-dis-crt-sd-new-yor/2335142.html
    Source snippet

    "CaseLaw[https://caselaw.findlaw.com/court/us-dis-crt-sd-new-yor/2335142.html..."](https://caselaw.findlaw.com/court/us-dis-crt-sd-new-yor/2335142.html...")...

  6. Source: responsibleailabs.ai
    Link: https://responsibleailabs.ai/ai-watch/mata-v-avianca-lawyers-sanctioned-for-chatgpt-fabricated-cases
    Source snippet

    Mata v. Avianca: Lawyers Sanctioned for ChatGPT-Fabricated Cases | AI Incident Watch | Responsible AI LabsJune 22, 2023...

    Published: June 22, 2023

Additional References

  1. Source: hinshawlaw.com
    Link: https://www.hinshawlaw.com/en/insights/lawyers-for-the-profession-alert/new-york-lawyer-who-cited-non-existent-cases-generated-by-chatgpt-now-faces-sanctions-hearing
    Source snippet

    York Lawyer Who Cited Non-Existent Cases Generated by ChatGPT Now Faces Sanctions Hearing | Hinshaw & Culbertson LLPMay 31, 2023...

    Published: May 31, 2023

  2. Source: reddit.com
    Link: https://www.reddit.com/r/Lawyertalk/comments/1qigg35/court_says_counsel_has_an_obligation_to_point_out/
    Source snippet

    says counsel has an obligation to point out AI hallucinations contained in opponent’s briefJanuary 20, 2026...

    Published: January 20, 2026

  3. Source: forbes.com
    Title: judge fines two lawyers for using fake cases from chatgpt
    Link: https://www.forbes.com/sites/mollybohannon/2023/06/22/judge-fines-two-lawyers-for-using-fake-cases-from-chatgpt/
    Source snippet

    "Forbes[https://www.forbes.com/sites/mollybohannon/2023/06/22/judge-fines-two-lawyers-for-using-fake-cases-from-chatgpt/..."](https://www.forbes.com/sites/mollybohannon/2023/06/22/judge-fines-two-lawyers-for-using-fake-cases-from-chatgpt...

  4. Source: reddit.com
    Title: www.reddit.com A I Misuse by Lawyers Isn’t Rare — It’s Just Being Measured Wrong
    Link: https://www.reddit.com/r/LawEthicsandAI/comments/1siv71d/ai_misuse_by_lawyers_isnt_rare_its_just_being/
    Source snippet

    Misuse by Lawyers Isn’t Rare — It’s Just Being Measured WrongApril 11, 2026...

    Published: April 11, 2026

  5. Source: reddit.com
    Title: www.reddit.com Trying to prevent Chat GPT from hallucinating legal reserach
    Link: https://www.reddit.com/r/ChatGPTPromptGenius/comments/1k6evyb
    Source snippet

    to prevent ChatGPT from hallucinating legal reserachApril 24, 2025...

    Published: April 24, 2025

  6. Source: legaldive.com
    Title: www.legaldive.com Lawyers citing fake Chat GPT cases are fined | Legal Dive
    Link: https://www.legaldive.com/news/lawyer-fake-chatgpt-cases-sanctions-fines-lawyers-chatgpt-fake-cases/653741/
    Source snippet

    citing fake ChatGPT cases are fined | Legal DiveJune 22, 2023...

    Published: June 22, 2023

  7. Source: youtube.com
    Title: Practice Law With AI (Without Losing Your Law License) | Justia Webinars
    Link: https://www.youtube.com/watch?v=jHQ1WpSQpVQ
    Source snippet

    Lawyer files ChatGPT DISASTER in COURT (Mata v. Avianca, Inc.)...

  8. Source: youtube.com
    Title: The Chat GPT Fake Cases Sanctions Decision (Mata v. Avianca)
    Link: https://www.youtube.com/watch?v=jPtOcsx7ATw
    Source snippet

    Practice Law With AI (Without Losing Your Law License) | Justia Webinars...

  9. Source: youtube.com
    Title: Lawyer files Chat GPT DISASTER in COURT (Mata v. Avianca, Inc.)
    Link: https://www.youtube.com/watch?v=GExSDY6Wz9Y
    Source snippet

    Relying on AI in the Practice of Law...

  10. Source: section1983.org
    Title: mata v avianca inc
    Link: https://www.section1983.org/cases/mata-v-avianca-inc/
    Source snippet

    v. Avianca, Inc. — Case Law Library | section1983.orgJune 22, 2023...

    Published: June 22, 2023

Topic Tree

Follow this branch

Parent topic

Avianca case The court case that exposed fake AI law

Related pages 2