Maintained by AI agents, supervised by legal engineers
legaltech.fyi Beta

Tracker

Sanctions watch

Every incident where a lawyer got caught filing AI-fabricated citations. The Hall of Shame.

Embed this
156
Incidents tracked
23
Incidents in 2026
$578,119
In fines, where reported
52 of 156 report an amount

The curve

Incidents per month

Last 18 months of published incidents, by the date of the court's order.

Mar 2025 Aug 2026

Charting 51 of 156 incidents — the rest are undated or fall outside this window.

Blamed by name

Counts only incidents where the filing court or the reporting named the tool.

Showing 156 incidents involving ChatGPT. Clear filters

Fined $2K ChatGPT
Cathleen Siebrecht filed a brief with AI-hallucinated citations and the court sanctioned her.
In re Cathleen Siebrecht · Cathleen Siebrecht · U.S. District Court for the Middle District of Florida · FL
“I absolve the mother of any intention to mislead the court.”

Cases that don't exist

  • Johnson v. City of Detroit, 2024 WL 1234567
peaceforasia.ch ↗
The file →
  • In re Cathleen Siebrecht; attorney Cathleen Siebrecht was sanctioned.
  • She filed a brief containing AI-hallucinated citations.
  • The court imposed a $2,500 monetary fine.
  • U.S. District Court for the Middle District of Florida; federal court in Florida.
  • Date not given on the page; the page links this as an AI hallucination case.
Caught, not sanctioned ChatGPT
They submitted a ChatGPT-produced legal brief to courts that cited fictitious cases.
Steven Schwartz and Peter LoDuca · Steven Schwartz and Peter LoDuca
mycase.com ↗
The file →
  • Steven Schwartz and Peter LoDuca were the lawyers identified on the page.
  • They presented a ChatGPT-produced brief with fictitious cases.
  • No sanction is described on the page, so the outcome is listed as none.
  • The page does not name a court or jurisdiction for the incident.
  • The page says they learned AI’s limitations the hard way and claimed ignorance.
Fined $5K ChatGPT
Jan Tomasik and Daniel Mann filed a brief with bogus citations generated by ChatGPT, and Judge Hardy ordered them removed, referred to the Nevada bar, and fined.
Uprise · Jan Tomasik; Daniel Mann; Cozen O’Connor · Washoe County, Nev., District Court · NV
“This is larger than any of us,”
lawnext.com ↗
The file →
  • Uprise; Jan Tomasik, Daniel Mann, and Cozen O’Connor
  • They filed a brief with bogus legal references generated by ChatGPT.
  • Judge Hardy ordered removal, bar referral, and $2,500 fines each ($5,000 total), then suspended them.
  • Washoe County, Nevada, District Court in Washoe County, Nev.
  • Last May; Hardy said he wanted system improvement and redemption.
Fined $3K ChatGPT
Walters filed papers with hallucinated citations; the court sanctioned him for the fake authorities.
Park v. Kim · Mark Walters · U.S. District Court, Southern District of New York · NY
“The opinion is short and direct.”

Cases that don't exist

  • Matter of Bourguignon v. Coordinated Behavioral Health Servs., Inc., 114 A.D.3d 947 (3d Dep’t 2014)
natlawreview.com ↗
The file →
  • Park v. Kim; Mark Walters was sanctioned
  • He filed briefs containing hallucinated legal citations
  • $3,000 monetary fine
  • U.S. District Court, Southern District of New York, New York
  • The page links the misconduct to ChatGPT-generated fake citations
Fined $10K ChatGPT
A California attorney filed ChatGPT-generated fake case quotes and citations, and the state appeals court fined the lawyer $10,000.
California attorney sanctioned for ChatGPT-generated fake case quotes and citations · state appeals court · CA
ai.plainenglish.io ↗
The file →
  • A California attorney was sanctioned for ChatGPT-fabricated case quotes.
  • The lawyer submitted 21 of 23 quotes that ChatGPT had completely fabricated.
  • The court fined the lawyer $10,000.
  • The sanction came from a state appeals court in California.
  • The page names ChatGPT as the source of the fake citations.
Caught, not sanctioned ChatGPT
A lawyer filed a brief with fabricated case law invented by ChatGPT, and the court sanctioned him after the error was uncovered.
ChatGPT Legal Case Incident
aihealthcarecompliance.com ↗
The file →
  • The page names "The ChatGPT Legal Case Incident" but gives no caption or lawyer name.
  • A lawyer submitted legal filings containing fabricated case law invented by ChatGPT.
  • The court sanctioned him after the error was uncovered.
  • No court, judge, date, or sanction amount is given on the page.
  • The page says Stanford HAI provides detailed analysis of the case.
Fined $5K ChatGPT
Schwartz filed a ChatGPT-written brief that cited nonexistent cases, and the court sanctioned him and fined him $5,000.
Steven A. Schwartz · Steven A. Schwartz · U.S. District Court for the Southern District of New York · NY
““I did not comprehend that ChatGPT could fabricate cases,” Schwartz later told the judge.”

Cases that don't exist

  • Martinez v. Delta
  • Zicherman v. Korean Air Lines
  • Varghese v. China Southern Airlines
precedent.com ↗
The file →
  • Steven A. Schwartz filed a ChatGPT-written brief in an airline suit.
  • The brief cited Martinez v. Delta, Zicherman v. Korean Air Lines, and Varghese v. China Southern Airlines.
  • He was sanctioned and fined $5,000.
  • The article places the matter in court in the Southern District of New York.
  • The cited cases did not exist, and Schwartz said he did not comprehend ChatGPT could fabricate cases.
Caught, not sanctioned ChatGPT
A lawyer submitted filings with AI-generated case citations; the BC Supreme Court declined special costs but held her personally liable for other costs.
Zhang v. Chen · counsel in Zhang v. Chen · British Columbia court · BC
“Citing fake cases in court filings and other materials handed up to the court is an abuse of process and is tantamount to making a false statement … Unchecked, it can lead to a miscarriage of justice.”
delcantochambers.com ↗
The file →
  • Zhang v. Chen involved a lawyer who submitted AI-generated case citations.
  • The filings contained hallucinated case law.
  • The BC Supreme Court declined special costs but held her personally liable for other costs.
  • The court was in British Columbia, Canada.
  • The article says the lawyer corrected errors, apologized, and did not intend to deceive.
Fined $10K ChatGPT
Two law firms faced sanctions after submitting ChatGPT-generated briefs that contained nine fabricated citations out of 27 total authorities.
ChatGPT-generated briefs
proplaintiff.ai ↗
The file →
  • Two law firms were sanctioned for ChatGPT-generated briefs.
  • Their filings contained nine fabricated citations out of 27 authorities.
  • The page says the sanction was a $10,000 wake-up call.
  • The court and judge are not named on the page.
  • The article gives the fabrication count but no citation strings.
Caught, not sanctioned ChatGPT
Two lawyers filed a ChatGPT-generated brief with fabricated court opinions and quotes, and a judge sanctioned them.
Two lawyers sanctioned in 2023
miamidadebar.org ↗
The file →
  • Two lawyers were sanctioned in 2023.
  • They filed a brief generated by ChatGPT with fabricated court opinions and quotes.
  • The page says a judge sanctioned them, but gives no amount.
  • No court, judge, or location is identified on the page.
  • The page cites ChatGPT and warns it can ‘hallucinate’ and be wholly inaccurate.
Caught, not sanctioned ChatGPT
A lawyer submitted a factum containing fictitious AI-generated case citations, and the court addressed the filing.
Ko v. Li · Ontario Superior Court of Justice · ON
“Ms. Lee forthrightly acknowledges the fact that her factum was created using ChatGPT and contains fake cases.”
canadianlawyermag.com ↗
The file →
  • Ko v. Li is named as a Canadian AI-citation incident.
  • A lawyer submitted a factum with fictitious AI-generated case citations.
  • The page does not state a sanction amount or type for this incident.
  • The court was in Ontario, Canada.
  • The page says the filings showed the risk of generative AI hallucinations.
Warned ChatGPT
A barrister misled the Tribunal by citing a fictitious Court of Appeal judgment generated by ChatGPT.
Upper Tribunal (Immigration and Asylum Chamber) · Upper Tribunal (Immigration and Asylum Chamber)
ein.org.uk ↗
The file →
  • Upper Tribunal (Immigration and Asylum Chamber) involved a barrister
  • The barrister cited a fictitious Court of Appeal judgment generated by ChatGPT
  • The Tribunal issued guidance and a stern warning
  • Upper Tribunal (Immigration and Asylum Chamber), England and Wales
  • Date not given; the page says this happened after ChatGPT-generated citations misled the Tribunal
Reprimanded ChatGPT
Three attorneys filed a motion with fabricated AI citations; a federal judge publicly reprimanded and disqualified them and referred them to the State Bar.
Alabama supervising attorney and others · federal judge · AL
“This was “an extreme dereliction of professional responsibility” and “lazy, convenient fictions” substituting for truth.”

Cases that don't exist

  • United States v. Baker
  • Kelley v. City of Birmingham
  • Greer v. Warden
vlex.com ↗
The file →
  • Alabama supervising attorney was one of three lawyers sanctioned over a motion with fake citations.
  • ChatGPT produced citations that were copied into the motion without verification.
  • The court publicly reprimanded and disqualified all three attorneys and referred them to the State Bar.
  • Federal court in Alabama.
  • Page says the judge called the conduct “recklessness in the extreme” and required review of 52 federal cases.
Fined $10K ChatGPT
An attorney filed AI-enhanced appellate briefs with fabricated citations and quotations; the California Court of Appeal sanctioned him $10,000 and referred him to the State Bar.
[Los Angeles attorney; case caption not named on page] · attorney in Los Angeles · California Court of Appeal · CA

Cases that don't exist

  • Schimmel v. Levin
  • Peake v. Underwood
vlex.com ↗
The file →
  • Los Angeles attorney was sanctioned after filing AI-enhanced appellate briefs.
  • ChatGPT generated fabricated citations and quotations that were filed without reading.
  • The court imposed a $10,000 sanction and referred him to the State Bar.
  • California Court of Appeal, California.
  • Page says the sanction was the highest fine ever issued by a California state court for AI misuse.
Caught, not sanctioned ChatGPT
Michael Long filed RTKL appeals and court papers with fabricated case citations, and the Borough challenged the filings in county court.
Michael Long · Michael Long · Northampton County Court of Common Pleas · PA

Cases that don't exist

  • Grove v. OOR
  • Pennsylvania State Education v. Commonwealth
  • IN RE S.S., 651 A.2d 174

+5 more

bathborough.org ↗
The file →
  • Michael Long was the requestor in the Bath Borough RTKL dispute.
  • He submitted legal filings and appeals containing fictitious or irrelevant citations.
  • The page reports no sanction order against him, only that the Borough challenged his filings.
  • The matter was in the Northampton County Court of Common Pleas in Pennsylvania.
  • The article says Long used AI and even bragged about using ChatGPT.
Caught, not sanctioned ChatGPT
A law school intern used ChatGPT for legal research, and an early-draft motion-to-dismiss opinion with fake quotes and misstated outcomes was withdrawn.
Julien Xavier Neals opinion · Julien Xavier Neals · U.S. District Court for the District of New Jersey · NJ
“In doing so, the intern acted without authorization, without disclosure, and contrary to not only chambers policy but also the relevant law school policy.”
abajournal.com ↗
The file →
  • Julien Xavier Neals, U.S. District Court for the District of New Jersey
  • A law school intern used ChatGPT and the opinion misstated case outcomes and used fake quotes.
  • The opinion was withdrawn; no sanction is stated on the page.
  • District of New Jersey; federal court.
  • Neals said the intern acted without authorization, without disclosure, and contrary to chambers and law school policy.
Warned ChatGPT
Owoc filed a court motion with 11 AI-hallucinated citations and was sanctioned by the court.
Jack Owoc matter · Jack Owoc · California court · CA
futurism.com ↗
The file →
  • Jack Owoc was sanctioned in a California case.
  • He filed a motion with 11 AI-hallucinated citations to nonexistent cases.
  • The court ordered 10 hours of community service and future AI disclosure.
  • California court; state jurisdiction; location not further specified on the page.
  • The article says he admitted using generative AI and the sanction was for hallucinated citations.
Warned ChatGPT
An immigration barrister cited entirely fictitious or irrelevant cases in appeal grounds and the Upper Tribunal judge said he had likely used generative AI and tried to hide it.
Chowdhury Rahman · Chowdhury Rahman · Upper Tribunal
“The most obvious explanation is … that the grounds of appeal were drafted in whole or in part by generative artificial intelligence such as ChatGPT.”
theguardian.com ↗
The file →
  • Chowdhury Rahman was the immigration barrister in the Upper Tribunal case.
  • He cited cases that were described as entirely fictitious or wholly irrelevant.
  • The judge said Rahman likely used generative AI and considered reporting him to the Bar Standards Board.
  • Court: Upper Tribunal in the UK; the ruling was published on Tuesday.
  • Judge Mark Blundell ruled in September and said the submissions wasted the tribunal’s time.
Reprimanded ChatGPT
Two outside attorneys cited cases that ChatGPT hallucinated in an unrelated lawsuit and were publicly reprimanded and referred to licensing authorities.
Baltimore Central Booking & Intake Center conditions case · Matthew Reeves and William Lunsford · U.S. District Court · MD
thebanner.com ↗
The file →
  • Matthew Reeves and William Lunsford were the attorneys.
  • They cited cases that ChatGPT hallucinated in an unrelated lawsuit.
  • Judge Anna M. Manasco publicly reprimanded them and referred them to bar authorities.
  • U.S. District Court in Alabama.
  • The article says this happened this year; the page names ChatGPT and the referral.
Fined $1K ChatGPT
Richard Bednar filed a brief with fake ChatGPT-generated citations, and the Utah Court of Appeals sanctioned him.
Richard Bednar · Richard Bednar · Utah Court of Appeals · UT
“"here is this case and it supports our position" and the case doesn’t exist.”

Cases that don't exist

  • Royer v. Nelson
  • Wadsworth v. Walmart
  • United States v. Hayes
cybernews.com ↗
The file →
  • Richard Bednar was sanctioned in the Utah Court of Appeals matter.
  • He submitted a brief with fake citations generated by ChatGPT.
  • He was ordered to pay fees, refund client fees, and donate $1,000.
  • Utah Court of Appeals in Utah.
  • The brief included the non-existent case Royer v. Nelson.
Fined $10K ChatGPT
A California attorney was fined for filing a state court appeal with fake quotations generated by ChatGPT.
California attorney sanctioned for fake quotations generated by ChatGPT · CA
cybernews.com ↗
The file →
  • A California attorney was sanctioned for an AI-generated filing.
  • The attorney used fake quotations generated by ChatGPT in a state court appeal.
  • The court imposed a $10,000 fine.
  • The matter was in California state court.
  • The page says the fine was issued in September and calls it historic.
Warned ChatGPT
A U.S. lawyer used ChatGPT to draft filings and cited fake cases; a federal judge responded with a standing order on AI use.
ChatGPT references nonexistent legal cases · A U.S. lawyer · federal judge
evidentlyai.com ↗
The file →
  • A U.S. lawyer in the ChatGPT case
  • He used ChatGPT and ended up citing entirely fake legal cases
  • The judge issued a standing order, not a monetary penalty
  • Federal court; the page gives no judge name or location
  • The page says the lawyer did not realize ChatGPT was a generative language tool
Caught, not sanctioned ChatGPT
Michael Cohen gave his lawyer AI-generated citations, and fake cases ended up in a real court filing.
Michael Cohen matter · Michael Cohen · U.S. District Court for the Southern District of New York · NY
“the attorneys’ actions were “embarrassing and certainly negligent.””
medium.com ↗
The file →
  • Michael Cohen and his lawyer were involved.
  • AI-generated citations were given to the lawyer and filed in court.
  • The page says they got sanctioned, but gives no order details.
  • No court, judge, or location is named on the page.
  • ChatGPT is identified as the tool used.
Caught, not sanctioned ChatGPT
Rose Lawyers drafted a will with AI-assisted legal citations; the page does not say any sanctions were imposed.
Re Walker [2025] VSC 714 · Rose Lawyers · Victorian Supreme Court
heirsandsuccesses.com ↗
The file →
  • Re Walker [2025] VSC 714 involved Rose Lawyers drafting the will.
  • The page discusses AI-assisted preparation and verification risks.
  • No sanction or penalty is described on the page.
  • Victorian Supreme Court, Victoria.
  • May 2024 guidelines and a discussion of ChatGPT/Google Gemini appear in the article.
Caught, not sanctioned ChatGPT
William Norman faced potential sanctions in two counties after ChatGPT created hallucinated quotes and left a prompt in a court document.
William Norman · William Norman
cleveland.com ↗
The file →
  • William Norman faced potential sanctions.
  • ChatGPT created hallucinated quotes and left a prompt in one court document.
  • The page reports potential sanctions but no imposed sanction amount.
  • The court and location are not named on the page.
  • The page says the issue involved two counties and ChatGPT.
Caught, not sanctioned ChatGPT
Relator’s expert used ChatGPT to draft an expert report with fabricated CMS testimony and bogus citations, and defendants moved to sanction counsel and seek disqualification.
United States ex rel. Khoury v. Mountain West Anesthesia, LLC · U.S. District Court for the District of Nevada · NV
“if fines and public embarrassment were effective deterrents, there would not be so many cases to cite.”
texaslawbook.net ↗
The file →
  • United States ex rel. Khoury v. Mountain West Anesthesia, LLC; relator’s counsel was at issue
  • An expert report used ChatGPT and contained fabricated CMS testimony, invented manuals and miscited regulations
  • Defendants moved for sanctions, fee shifting and disqualification; no sanction order is stated on the page
  • U.S. District Court for the District of Nevada, federal court in Nevada
  • Aug. 13 motion; the page says the misconduct may lead to disqualification and dismissal
Referred to the bar ChatGPT
A lawyer was referred to a disciplinary regulator after filing two documents that relied on non-existent citations and quotes.
ChatGPT blunder sees lawyer referred to regulator · Disciplinary regulator
lawyersweekly.com.au ↗
The file →
  • A lawyer was referred to a disciplinary regulator.
  • The page says two filed documents relied on non-existent citations and quotes.
  • The sanction was referral; no amount is stated.
  • The article names ChatGPT as the AI tool.
  • Published Dec. 31, 2025; the page gives no case caption or judge quote.
Caught, not sanctioned ChatGPT
Counsel was said to have filed a letter brief with fabricated AI citations, and Judge Ruth Bermudez Montenegro dismissed the case for lack of diversity jurisdiction.
LaPaglia v. Valve Corp. · S.D. Cal. · CA

Cases that don't exist

  • Bassett's Adm'r v. Cunningham's Adm'r, 50 Va. 684
reason.com ↗
The file →
  • LaPaglia v. Valve Corp. involved petitioner's counsel.
  • Valve said counsel submitted a letter brief with AI-fabricated citations.
  • Judge Ruth Bermudez Montenegro dismissed the case for lack of diversity jurisdiction.
  • S.D. Cal., federal court in California.
  • The page says the counsel had been recently sanctioned by an arbitrator; ChatGPT was used only in a law-clerk check.
Fined $13K ChatGPT
A federal judge sanctioned Hagens Berman Sobol Shapiro, Robert Carey, and Celeste Boyd after briefs in an OnlyFans case contained AI-generated hallucinated citations.
OnlyFans users v. Fenix International Ltd. · Hagens Berman Sobol Shapiro; Robert Carey; Celeste Boyd · U.S. District Court for the Central District of California · CA
“He wrote that allowing revisions would unfairly burden defendants who had already responded, noting that the proposed corrections still contained errors.”
complex.com ↗
The file →
  • OnlyFans users v. Fenix International Ltd. involved Hagens Berman Sobol Shapiro, Robert Carey, and Celeste Boyd.
  • Four briefs reportedly contained fabricated or hallucinated legal citations generated through AI tools.
  • Judge Fred Slaughter imposed $10,000 on the firm and Carey, plus $3,000 on Boyd.
  • U.S. District Court for the Central District of California; Seattle-based firm; California federal case.
  • The article says Boyd used ChatGPT to draft and edit sections and failed to independently verify the output.
Fined $500 ChatGPT
Michael Chambers filed a motion containing ChatGPT-generated fake case citations, and the court sanctioned him for the AI-made fabrications.
Arthur v. City of Hartford · Michael Chambers · U.S. District Court for the District of Connecticut · CT

Cases that don't exist

  • Mati v. City of Hartford
  • Smith v. City of Hartford
  • Johnson v. City of Hartford
nytimes.com ↗
The file →
  • Arthur v. City of Hartford; Michael Chambers was sanctioned.
  • He filed a motion with ChatGPT-generated fake legal citations.
  • The court imposed a $500 monetary fine.
  • U.S. District Court for the District of Connecticut.
  • The page says the fake cases came from ChatGPT.
Caught, not sanctioned ChatGPT
A practitioner filed a submission quoting a fictional Court of Appeal judgment generated by ChatGPT, and the judge said it wasted public resources.
LJY v Occupational Therapy Board of Australia · Queensland Civil and Administrative Tribunal · QLD
“the case simply does not exist”
nicholeslaw.com.au ↗
The file →
  • LJY v Occupational Therapy Board of Australia; a practitioner was involved.
  • A written submission quoted a fictional Court of Appeal judgment generated by ChatGPT.
  • No sanction is stated on the page; the judge said it wasted public resources.
  • Queensland Civil and Administrative Tribunal, Queensland.
  • 2025; the judge said 'the case simply does not exist' and 'caused a significant waste of public resources.'
Filing dismissed ChatGPT
A worker filed AI-generated WCAT submissions with fabricated cases and policies; WCAT dismissed the appeal and warned of possible costs.
A2501051 · Workplace Safety and Compensation Appeal Tribunal
“It appears that the worker’s submission was created, at least partly, with the use of artificial intelligence.”

Cases that don't exist

  • Zhang v. Chen, 2024 BCSC 285
  • Geismayr v. The Owners, Strata Plan KAS 1970, 2025 BCCRT 217
  • AQ v. BW, 2025 BCCRT 907

+1 more

wcblawyers.com ↗
The file →
  • A2501051 at WCAT involved a worker's submission.
  • The worker used AI to prepare arguments for a prohibited action appeal.
  • The submission contained fake cases, incorrect policy, and bad legal arguments.
  • WCAT dismissed the appeal and noted AI-based submissions can lead to costs.
  • The deputy registrar said the submission was partly created with artificial intelligence and cited hallucinated cases.
Caught, not sanctioned ChatGPT
An attorney used ChatGPT to draft a brief that cited made-up cases, and the resulting case was later recognized as fabricated.
Varghese v. China Southern Airlines

Cases that don't exist

  • Varghese v. China Southern Airlines
zdnet.com ↗
The file →
  • Varghese v. China Southern Airlines involved an attorney who used ChatGPT to prepare a legal brief.
  • The brief contained legal cases that did not actually exist.
  • The page does not name a sanction in this incident.
  • The article says the problem case was fabricated by ChatGPT.
  • The page gives no judge, court, or sanction date for this incident.
Caught, not sanctioned ChatGPT
A litigant in person filed a submission using a legal phrase that looked Scottish or American, and the judge said it had the hallmarks of ChatGPT.
Reddan v An Bord Pleanála · High Court of Ireland
“This sounds like something that derived from an artificial intelligence source. It has all the hallmarks of ChatGPT, or some similar AI tool.”
fastdatascience.com ↗
The file →
  • Reddan v An Bord Pleanála involved a litigant in person.
  • The filing used a phrase not used in Ireland and suggested AI generation.
  • No sanction is described on the page.
  • High Court of Ireland.
  • The judge said it had the hallmarks of ChatGPT or similar AI.
Caught, not sanctioned ChatGPT
A litigant in person used ChatGPT to draft an appeal submission containing fake cases, and the ruling reiterated that an unrepresented person must not mislead the court.
Barton v Wright Hassell LLP · UK Supreme Court
““an unrepresented person is still under a duty not to mislead the court””
fastdatascience.com ↗
The file →
  • Barton v Wright Hassell LLP involved a litigant in person.
  • He used ChatGPT to draft a submission with fake cases.
  • The page reports no sanction, only a warning principle from the ruling.
  • UK Supreme Court.
  • The ruling quoted the duty of an unrepresented person not to mislead the court.
Caught, not sanctioned ChatGPT
Attorneys signed a brief containing AI-hallucinated legal citations and had to respond to a judge's order to explain why they should not be punished.
Jason Alatidd article (Kansas federal court brief) · federal court in Kansas · KS
cjonline.com ↗
The file →
  • Attorneys who signed a Kansas federal-court brief
  • The brief contained AI-hallucinated legal citations
  • They responded to a judge's order to explain why they should not be punished
  • Federal court in Kansas
  • Jan. 8, 2026 article; the AI tool named was ChatGPT
Caught, not sanctioned ChatGPT
The Upper Tribunal held a Hamid hearing over two immigration matters involving incorrect or fake case citations that did not match the cited cases.
UK v Secretary of State for the Home Department (AI hallucinations; supervision; Hamid) [2026] UKUT 81 (IAC) · Upper Tribunal (Immigration and Asylum Chamber)
“to put client letters and decision letters from the Home Office into an open source AI tool, such as ChatGPT, is to place this information on the internet in the public domain, and thus to breach client confidentiality and waive legal privilege, and thus any regulated legal professional or firm that does so would, in addition to needing to bring this to the attention of their regulator, be advised to consult with the Information Commissioner’s Office.”
gowlingwlg.com ↗
The file →
  • UK v Secretary of State for the Home Department (AI hallucinations; supervision; Hamid) [2026] UKUT 81 (IAC) was the latest case discussed.
  • It involved two separate instances of suspected AI-invented or otherwise incorrect case citations.
  • The article says one case looked trivial and the other serious, but the tribunal may not have actually sanctioned anyone.
  • The court was the Upper Tribunal (Immigration and Asylum Chamber).
  • Published 2026-02-24; the article notes the tribunal may have been too quick to cry AI.
Referred to the bar ChatGPT
Two judicial review applications contained false case citations generated by non-specialist AI tools, and the Tribunal warned that supervisors and signatories risk referral.
R (on the application of Munir) v Secretary of State for the Home Department · Upper Tribunal (Immigration and Asylum Chamber)
“Legal professionals are obliged to ensure that legal arguments which are presented to the First-tier Tribunal or Upper Tribunal are factually and legally accurate.”
ein.org.uk ↗
The file →
  • R (on the application of Munir) v Secretary of State for the Home Department involved two judicial review applications.
  • The grounds and supporting material contained false and non-existent case citations from AI drafting.
  • The Tribunal warned failures to check work are likely to result in referral to the SRA or another regulator.
  • Upper Tribunal (Immigration and Asylum Chamber), UK.
  • Published 2026-02-20; the page says ChatGPT was one open-source AI tool discussed.
Warned ChatGPT
Three lawyers were warned after using ChatGPT to cite fake or incorrect legal cases in court, and two were ordered to take an AI training course.
Dutch lawyers warned for misusing ChatGPT in court · Three lawyers in the Netherlands
nltimes.nl ↗
The file →
  • Three lawyers in the Netherlands were warned
  • They misused ChatGPT to cite fake or incorrect legal cases in court
  • Sanction was a warning, and two lawyers had to take an AI training course
  • No court or judge name is given in the page text
  • The report says two were ordered to take an AI training course
Referred to the bar ChatGPT
A solicitor submitted a filing with a fake case name and false citations; the Upper Tribunal said it would have referred him to regulators but for his self-reporting.
Tahir Mehmood Mohammed · Tahir Mehmood Mohammed · Upper Tribunal, Immigration and Asylum Chamber
“It matters not how such citation errors come about.”
legalfutures.co.uk ↗
The file →
  • Tahir Mehmood Mohammed, a solicitor at TMF Immigration Lawyers, filed false citations.
  • He said he may have used Google AI mode and had also put client material into ChatGPT.
  • The tribunal would have referred him to the IAA and SRA but for self-reporting.
  • Upper Tribunal, Immigration and Asylum Chamber; judge Fiona Lindsley.
  • November 2025 decision, published on 24 February 2026; the judge said errors matter regardless of whether they came from a trainee or ChatGPT.
Warned ChatGPT
A firm filed two briefs with five false ChatGPT-generated citations, and a federal judge was considering sanctions and potential fines.
another firm · federal judge
courtroomsciences.com ↗
The file →
  • Another firm faced possible sanctions for false citations.
  • Attorneys filed two briefs containing five ChatGPT-generated false citations.
  • The judge was considering sanctions and potential fines; no amount is given.
  • A federal judge in federal court.
  • The partner admitted using ChatGPT without verification and the firm apologized.
Fined ChatGPT
A Utah attorney filed false ChatGPT-generated citations, including Royer v. Nelson, and the state court of appeals sanctioned the attorney.
Royer v. Nelson · state court of appeals · UT

Cases that don't exist

  • Royer v. Nelson
courtroomsciences.com ↗
The file →
  • A Utah attorney was sanctioned in Royer v. Nelson.
  • The filing contained a nonexistent case and other false citations generated by ChatGPT.
  • The court imposed sanctions; the article does not state the amount.
  • State court of appeals in Utah.
  • 2025; opposing counsel said the material was unverifiable in any legal database.
Fined ChatGPT
Steven A. Schwartz filed a brief with AI-fabricated cases and quotations, and the court sanctioned the filing.
In re Schwartz · Steven A. Schwartz · federal court
prinz-lawfirm.com ↗
The file →
  • In re Schwartz involved Steven A. Schwartz and AI-fabricated citations.
  • He submitted a brief with made-up cases and quotations.
  • The court sanctioned the filing with a monetary fine.
  • The page says courts in Wyoming, California, Texas, and Pennsylvania have sanctioned similar conduct.
  • It identifies ChatGPT as one of the AI tools that can hallucinate citations.
Fined ChatGPT
Steven Schwartz was sanctioned after filing a brief with six AI-generated case citations that were completely fabricated.
Steven Schwartz · Steven Schwartz · S.D.N.Y. · NY
law360.com ↗
The file →
  • Steven Schwartz was sanctioned in a 2023 New York case.
  • He submitted a brief with six AI-generated case citations.
  • The page says the citations were completely fabricated.
  • The court imposed sanctions in federal court in the Southern District of New York.
  • The page does not name the AI tool or give the sanction amount.
Fined $2K ChatGPT
An attorney filed a pleading citing at least six nonexistent cases, and Judge Martínez fined him $1,500 and ordered ethics training and bar reporting.
Unnamed New Mexico attorney · unnamed New Mexico attorney · U.S. Magistrate Judge Damian Martínez · NM
“The six cases were fake and likely the handiwork of a ChatGPT or similar artificial intelligence (AI) program’s hallucinations,”
govtech.com ↗
The file →
  • An unnamed New Mexico attorney was sanctioned for fake citations.
  • He filed a pleading with at least six nonexistent cases.
  • Judge Martínez fined him $1,500 and required bar reporting and AI ethics training.
  • U.S. Magistrate Judge Damian Martínez, Las Cruces, New Mexico.
  • Last year; the article says the fake cases were likely from ChatGPT or a similar AI program.
Warned ChatGPT
A lawyer’s suspected AI use led to fake cases being cited before the tribunal, and the tribunal warned that uploading confidential client documents to ChatGPT breached privilege.
Munir v. Secretary of State for the Home Department · Immigration and Asylum Chamber
blakes.com ↗
The file →
  • Munir v. Secretary of State for the Home Department involved a lawyer suspected of using AI-generated fake cases.
  • The lawyer said he did not know how the fake cases appeared in his submissions.
  • The tribunal warned that uploading confidential client documents into an open-source AI tool breaches client confidentiality and waives privilege.
  • Immigration and Asylum Chamber (U.K.).
  • The page says the lawyer admitted putting client letters and other confidential material in ChatGPT.
Fined $6K ChatGPT
Debtor’s counsel filed a brief with AI-generated fabricated citations and nonexistent quotations, and the bankruptcy court sanctioned counsel and his firm.
In re Marla C. Martin · U.S. Bankruptcy Court for the Northern District of Illinois · IL
““no lawyer should be using … a generative AI product to perform research without verifying results. Period.””
duanemorris.com ↗
The file →
  • Marla Martin bankruptcy case; debtor’s counsel and his law firm were sanctioned
  • Counsel filed a brief with fabricated citations and nonexistent quotations generated by AI
  • The court imposed a $5,500 penalty and required AI-training attendance
  • U.S. Bankruptcy Court for the Northern District of Illinois, Illinois
  • November 7, 2025 article; the court called the sanction "modest" and warned future violators
Fined $1K ChatGPT
Two New Orleans law department attorneys filed a motion with nine AI-generated fake case citations, and Judge Carl Barbier sanctioned them.
Jalen Harris and James Roquemore · Assistant City Attorney Jalen Harris; Deputy City Attorney James Roquemore · U.S. District Court for the Eastern District of Louisiana · LA
abajournal.com ↗
The file →
  • Assistant City Attorney Jalen Harris and Deputy City Attorney James Roquemore were sanctioned in federal court.
  • They filed a January 2026 motion containing nine made-up case citations generated by AI.
  • Judge Carl Barbier ordered Harris to pay $250 and Roquemore $1,000.
  • The case was in the U.S. District Court for the Eastern District of Louisiana in New Orleans.
  • Harris said he used ChatGPT to speed research and did not verify the AI-generated cases.
Fined $5K ChatGPT
Two New York lawyers filed a brief with six fake ChatGPT-created case citations and were fined $5,000.
Two New York lawyers sanction incident · two New York lawyers · United States District Court for the Southern District of New York · NY
cnbc.com ↗
The file →
  • Two New York lawyers were the sanctioned attorneys.
  • They submitted a legal brief with six fake case citations created by ChatGPT.
  • The court fined them $5,000.
  • The matter was in federal court in New York.
  • The page says this happened in 2023, and the AI tool was ChatGPT.
Caught, not sanctioned ChatGPT
Plaintiffs' attorneys filed AI-generated bogus case law, and the court issued an order to show cause why they should not be sanctioned.
Order to Show Cause Why Plaintiffs' Attorneys Should Not Be Sanctioned · United States District Court for the District of Wyoming · WY
lib.law.uw.edu ↗
The file →
  • Plaintiffs' attorneys in the District of Wyoming
  • They submitted bogus case law created by ChatGPT
  • The court issued an order to show cause, with sanctions threatened
  • United States District Court for the District of Wyoming
  • 2025; the page links this to a 2025 order and names Judge Kelly H. Rankin
Fined $5K ChatGPT
A New York lawyer filed court papers citing fake legal cases generated by ChatGPT, and the judge fined the lawyer and his firm $5,000.
Unknown New York lawyer matter · a New York lawyer · court not named on page · NY
bloomfire.com ↗
The file →
  • A New York lawyer and his firm were fined in court.
  • He filed papers with fake legal cases generated by ChatGPT.
  • The judge imposed a $5,000 fine.
  • The page does not name the court or judge.
  • The article names ChatGPT as the source of the fabricated citations.
Warned ChatGPT
Two lawyers filed a brief with non-existent cases; the High Court warned of severe sanctions after ChatGPT was found to have been used.
High Court of Justice in London warning matter · High Court of Justice in London
“The technology was “not capable of conducting reliable legal research.””
posteo.de ↗
The file →
  • Two lawyers submitted a brief citing cases that could not be found.
  • One lawyer had asked ChatGPT for evidence supporting the claim.
  • The High Court warned of severe sanctions, not a monetary penalty.
  • High Court of Justice in London.
  • Last summer; judge said the tools can produce plausible but incorrect responses.
Outcome unclear ChatGPT
Mr. Sauve filed a new federal complaint using ChatGPT and Claude that was described as a fake case littered with hallucinated citations in violation of court rules.
Sauve · Mr. Sauve · federal court
nytimes.com ↗
The file →
  • Mr. Sauve filed a new federal complaint.
  • The pleading used ChatGPT and Claude and was described as a fake case.
  • The page does not state any sanction or amount.
  • The court is identified only as federal court.
  • No date or judge name is given; the article says the complaint had hallucinated citations.
Caught, not sanctioned ChatGPT
Attorney Harp was ordered to produce his ChatGPT history so the court could determine whether he had submitted hallucinated AI quotations.
Attorney Harp · Attorney Harp · United States District Court, Northern District of California · CA
websitedc.s3.amazonaws.com ↗
The file →
  • Attorney Harp was the lawyer under scrutiny.
  • The issue was whether he submitted hallucinated AI quotations to the court.
  • The court ordered him to produce his ChatGPT history.
  • United States District Court, Northern District of California.
  • No sanction amount was stated; the order was to investigate possible AI misuse.
Fined $6K ChatGPT
The firm submitted a brief containing generative-AI mistakes and erroneous citations, and a state appellate panel upheld $6,000 in sanctions.
Lipeles Law Group sanctions appeal · Lipeles Law Group · California appellate panel · CA
“the worst example of misconduct by a lawyer that I think I’ve ever seen since I’ve been on the bench.”
law.com ↗
The file →
  • Lipeles Law Group was sanctioned after filing a brief with AI mistakes and erroneous cites.
  • The article says a contract attorney was responsible for the citation errors.
  • The appellate panel upheld $6,000 in sanctions.
  • A California state appellate panel handled the matter.
  • The article says the attorney used a 'premier corporate subscription of ChatGPT'.
Warned ChatGPT
A regulated immigration adviser filed an appeal application using fictitious case law from Google’s AI Mode and faced potential SRA censure.
UK v Secretary of State for the Home Department · Mr. Zubair Rasheed of City Law Practice Solicitors · Upper Tribunal (Immigration and Asylum Chamber)
“it would be easy to think that this is a case about the naïve use of generative AI, but it is not merely about that; it is principally about supervision and the obligation to ensure that the tribunal is not misled.”
thelens.slaughterandmay.com ↗
The file →
  • UK v Secretary of State for the Home Department involved a regulated immigration adviser.
  • He used fictitious case law from Google’s AI Mode in an application for permission to appeal.
  • The court said it would have reported him if he had not self-reported.
  • Court: Upper Tribunal (Immigration and Asylum Chamber), UK.
  • The adviser also put draft emails and Home Office letters into ChatGPT.
Caught, not sanctioned ChatGPT
Farah Chalisa said she used ChatGPT for stylistic and grammatical review, and the court discussed the resulting AI-related citation problem.
Salah v. Walmart Inc. · Farah Chalisa · U.S. Court of Appeals for the Seventh Circuit
media.ca7.uscourts.gov ↗
The file →
  • Salah v. Walmart Inc.; Farah Chalisa
  • Chalisa said she used ChatGPT for stylistic and grammatical review
  • The page does not state any sanction or amount for this order
  • U.S. Court of Appeals for the Seventh Circuit
  • The page names ChatGPT and says Chalisa did not enter an appearance
Reprimanded ChatGPT
Forrister submitted AI-fabricated citations in a filing and faced discipline from the Texas Supreme Court.
In re: The State Bar of Texas v. Christopher W. Forrister · Christopher W. Forrister · Supreme Court of Texas · TX
caseread.ai ↗
The file →
  • Christopher W. Forrister was disciplined for AI-fabricated citations.
  • He filed a document that included fake citations generated by AI.
  • The Texas Supreme Court issued a reprimand.
  • The court was the Supreme Court of Texas.
  • The page names ChatGPT as the AI tool but gives no order date.
Caught, not sanctioned ChatGPT
Michael Best was cited as an example of a lawyer or firm facing sanction risk for submitting AI-generated filings with fabricated citations.
Michael Best · Michael Best
lexology.com ↗
The file →
  • Michael Best was named on the page as a real incident reference.
  • The page says it involved AI-generated filings with fabricated citations.
  • No specific sanction amount or order is stated on the page.
  • No court, judge, or location is identified on the page.
  • The page links the problem to ChatGPT and fabricated legal guidance.
Caught, not sanctioned ChatGPT
A relator’s expert submitted a report with AI-generated hallucinations, including fabricated government testimony and fake citations, and the relator withdrew the expert after the court was alerted.
U.S. ex rel. Khoury v. Mountain West Anesthesia LLC · U.S. District Court for the District of Utah · UT
bassberry.com ↗
The file →
  • Mountain West Anesthesia FCA case and one of the relator’s experts
  • The expert’s report contained AI-generated hallucinations, including fake citations
  • The court was alerted and the relator withdrew the expert; sanctions were sought but not imposed on the page
  • Utah federal court, Utah
  • September 2025; the expert acknowledged using ChatGPT
Caught, not sanctioned ChatGPT
A Denver attorney accepted a 90-day suspension after an investigation found he used ChatGPT to help draft a motion containing fabrications.
Unnamed Colorado disciplinary case · Denver attorney · Colorado Supreme Court · CO
“like an idiot”
cronkitenews.azpbs.org ↗
The file →
  • Unnamed Colorado disciplinary case; a Denver attorney was sanctioned.
  • The attorney denied using AI, but the investigation found ChatGPT helped draft a motion with fabrications.
  • He received a 90-day suspension.
  • Colorado Supreme Court; Colorado.
  • The article gives no date beyond saying the case was recent; it quotes the attorney saying he hadn’t checked the work.
Fined $10K ChatGPT
An attorney filed a state court appeal with 21 of 23 quoted citations hallucinated by ChatGPT and was fined $10,000.
California court appeal matter · California court · CA
nbcnews.com ↗
The file →
  • An unnamed attorney in California was fined.
  • The attorney filed a state court appeal with 21 of 23 quotes hallucinated by ChatGPT.
  • The court imposed a $10,000 fine.
  • California court; state jurisdiction; location not further specified on the page.
  • The article calls it possibly the largest-ever fine over AI fabrications.
Caught, not sanctioned ChatGPT
Takefman cited the nonexistent case Hernandez v. Gilbert in two filings, but the court accepted his apology and did not sanction him.
Takefman matter · Earl Takefman · court not named on page · FL

Cases that don't exist

  • Hernandez v. Gilbert
nbcnews.com ↗
The file →
  • Earl Takefman was the pro se litigant in a Florida matter.
  • He used ChatGPT and cited the imaginary case Hernandez v. Gilbert in two filings.
  • The court forced him to explain why he should not be sanctioned, then imposed none.
  • Florida; court not named on the page; state jurisdiction.
  • The page says the judge admonished him and later accepted his apology.
Fined $10K ChatGPT
An appeals court in California fined a lawyer $10,000 after nearly all quotations in the briefs were fabricated with AI tools.
In re: OpenAI, Inc. ChatGPT Litigation · California Court of Appeal · CA
““as a warning””
404media.co ↗
The file →
  • An unnamed lawyer was sanctioned in a California appeal.
  • The lawyer filed briefs with nearly all legal quotations fabricated through AI tools.
  • The court imposed a record $10,000 fine.
  • California Court of Appeal, California.
  • The page says the lawyer used ChatGPT, Claude, Gemini, and Grok; the date is only given as earlier this month.
Fined ChatGPT
An attorney and his firm were fined after filing bankruptcy arguments containing four misrepresented or fictitious matters.
Bankruptcy proceedings matter · United States Bankruptcy Court
heplerbroom.com ↗
The file →
  • An attorney and his firm were sanctioned in bankruptcy proceedings.
  • Their filing cited four misrepresented or fictitious matters.
  • The court fined both the attorney and the firm.
  • United States Bankruptcy Court; the article gives no judge name.
  • The attorney admitted using ChatGPT and not verifying the citations before filing.
Fined $60K ChatGPT
Counsel cited the nonexistent Illinois Supreme Court case Mack v. Anderson in a post-trial motion; the judge held a special hearing and sanctions were sought.
Chicago Housing Authority matter · Chicago Housing Authority's counsel · Cook County Circuit Court · IL

Cases that don't exist

  • Mack v. Anderson
heplerbroom.com ↗
The file →
  • Chicago Housing Authority counsel cited Mack v. Anderson in a post-trial motion.
  • The case did not exist and was generated by AI research.
  • Plaintiff sought sanctions after the false citation and related misrepresentations.
  • Cook County Circuit Court, Illinois; Judge Thomas Cushing held a special hearing.
  • Summer 2025; the attorney said she did not think ChatGPT could create false precedent.
Fined $500 ChatGPT
The plaintiff filed AI-prepared pleadings with fictitious legal authorities and was ordered to pay $500 in compensation.
Kalala c. Coopérative d'habitation La Fraternité Micheloise · Superior Court of Québec
lazaruslegal.com ↗
The file →
  • Kalala c. Coopérative d'habitation La Fraternité Micheloise; the plaintiff
  • The pleadings contained fictitious authorities and wrong Civil Code provisions
  • Monetary sanction of $500 in compensation
  • Superior Court of Québec
  • 2025; the plaintiff acknowledged using ChatGPT
Fined $15K ChatGPT 2026-07-28
Mason Cole filed briefs with false statutory quotations and case citations, and the Illinois Appellate Court for the First District fined him for each false citation and quotation.
Mason Cole · Mason Cole of Cole Sadkin · Illinois Appellate Court for the First District · IL
“"This is unacceptable," the appeals court said.”
abajournal.com ↗
The file →
  • Mason Cole of Cole Sadkin was sanctioned.
  • He filed briefs containing false statutory quotations, false case citations and improper facts.
  • The court imposed a $15,000 fine, $1,500 for each false citation and quotation.
  • Illinois Appellate Court for the First District, Illinois.
  • July 28 opinion; Cole said he used a premium corporate subscription of ChatGPT and cross-referenced with LexisNexis.
Fined $3K ChatGPT 2026-07-24
A plaintiff's attorney filed multiple court submissions with dozens of AI-generated fake citations and quotations and did not correct them; the judge imposed a $3,000 sanction.
Joann LeDoux v. Outliers, Inc. · U.S. District Court for the Western District of Washington · WA
“tantamount to bad faith”
law360.com ↗
The file →
  • A plaintiff's attorney in Tiffany M. Cartwright was sanctioned.
  • She submitted multiple filings with dozens of fake citations and quotations generated by AI platforms.
  • The court imposed a $3,000 monetary sanction.
  • U.S. District Court for the Western District of Washington, Washington.
  • July 24, 2026; the judge called the conduct 'tantamount to bad faith.'
Suspension $10K ChatGPT 2026-07-17
The tribunal temporarily suspended Lee’s law license after finding she filed a factum with AI-generated non-existent case law and lied about using ChatGPT.
Mary Hyun-Sook Lee · Mary Hyun-Sook Lee (also known as Jisuh Lee) · Law Society Tribunal · ON
“It should go without saying that it is the lawyer’s duty to read cases before submitting them to a court as precedential authorities.”
lawtimesnews.com ↗
The file →
  • Mary Hyun-Sook Lee, also known as Jisuh Lee, was temporarily suspended by the Law Society Tribunal.
  • She relied on a factum containing non-existent or irrelevant case law generated by an AI tool.
  • The tribunal also ordered her to pay $10,000 to the Law Society of Ontario.
  • The matter was before the Law Society Tribunal in Ontario, Canada.
  • On July 17, 2026, the tribunal said she lied about not using ChatGPT and had failed to act with honour and integrity.
Reprimanded ChatGPT 2026-07-07
The Connecticut Supreme Court found seven ChatGPT-caused filing errors by a GLG Law LLC attorney and ordered continuing legal education, finding negligence rather than intent to mislead.
TOV Realty, LLC v. Angel Suarez et al. / Kosel Equity, LLC v. Mark MacGregor et al. · Ian G. Gottlieb; GLG Law LLC · Connecticut Supreme Court · CT
“why they should not be sanctioned for misrepresenting “the law through the use of generative artificial intelligence.””
law360.com ↗
The file →
  • GLG Law LLC attorney Logan Vance was the subject of the order.
  • Seven filing errors were caused by ChatGPT, according to the court.
  • The court ordered continuing legal education, not money.
  • Connecticut Supreme Court, Connecticut.
  • August 3, 2026; the court found negligence in using new technology, not an intent to deceive.
Fined ChatGPT 2026-07-02
An Illinois attorney admitted using ChatGPT in a response brief filed July 2, and the court said the brief also contained misstatements of law.
Ill. Appeals Court Says It Hopes Higher Fine Will Deter AI Hallucinations · IL
law.com ↗
The file →
  • Illinois appeals-court matter about hallucinated citations.
  • The attorney admitted using a 'premier corporate subscription of ChatGPT' in a response brief.
  • The court said the brief also included misstatements of law.
  • Illinois Appeals Court.
  • The response brief was filed July 2, 2026; the article says the court hoped a higher fine would deter AI hallucinations.
Fined $6K ChatGPT 2026-07-01
Thomas Nield used ChatGPT to research a Chapter 13 issue and filed briefs with nonexistent cases and quotes; Judge Michael Slade fined him $5,500.
Thomas Nield · Thomas Nield · Illinois bankruptcy court · IL
“The next lawyer who does the same thing is warned that he or she will likely see a more significant penalty.”
news.bloomberglaw.com ↗
The file →
  • Thomas Nield, a former Semrad Law Firm bankruptcy attorney, was sanctioned in Illinois.
  • He used ChatGPT and filed documents with hallucinated cases and quotes.
  • Judge Michael Slade imposed a $5,500 fine.
  • The sanction came from the Illinois bankruptcy court.
  • The article says he used ChatGPT for an ancillary issue in a Chapter 13 case and was terminated in September.
Reprimanded ChatGPT 2026-05-21
H. Gregory Harp filed a response with false quotations attributed to real cases and was sanctioned by Judge Harold D. Mooty III.
Jackie Miller v. Regions Bank · H. Gregory Harp · U.S. District Court for the Northern District of Alabama · AL
““Lawyers make errors. Competent and ethical lawyers own them””
findlaw.com ↗
The file →
  • Jackie Miller v. Regions Bank; attorney H. Gregory Harp
  • Harp filed a response with false quotations in support of summary judgment briefing
  • He was sanctioned and suspended for six months in the Northern District of Alabama
  • U.S. District Court for the Northern District of Alabama, Alabama
  • May 21, 2026; the court suspected generative AI and Harp later admitted using free ChatGPT for proofreading
Warned ChatGPT 2026-05-05
Kelly Guagenty filed a brief with AI-caused citation errors, mischaracterizations and an improper quotation; Judge Stacey D. Neumann ordered AI training and new safeguards but no fine.
Jessica Fuller v Hyde School · Karen Guagenty · U.S. District Court for the District of Maine · ME
“Although AI can be a useful aid in research and drafting, its use does not diminish an attorney's nondelegable duties of diligence, candor, and reasonable inquiry.”
habeas.ai ↗
The file →
  • FULLER v. HYDE SCHOOL involved Attorney Kelly Guagenty.
  • She submitted two filings with AI-generated erroneous legal citations.
  • The court imposed non-monetary sanctions.
  • United States District Court for the District of Maine, Maine.
  • May 5, 2026; counsel said the AI used was either Claude or ChatGPT.
Caught, not sanctioned ChatGPT 2026-04-22
Sullivan & Cromwell filed an emergency motion with AI hallucinations and inaccurate citations, then apologized and corrected the filing before any sanctions order.
In re: Prince Global Holdings Ltd. · Sullivan & Cromwell LLP · U.S. Bankruptcy Court for the Southern District of New York · NY
reuters.com ↗
The file →
  • In re: Prince Global Holdings Ltd.; Sullivan & Cromwell LLP
  • A court submission contained fabricated or misquoted case citations from AI hallucinations
  • No sanction is stated in the article; the firm apologized
  • U.S. Bankruptcy Court for the Southern District of New York
  • 2026-04-18; the article says the filing was made on April 18 and involved AI hallucinations
Caught, not sanctioned ChatGPT 2026-03-23
A solicitor filed appeal grounds with a fake citation and later admitted likely using Google’s AI mode and ChatGPT; the Upper Tribunal criticized the conduct but imposed no sanction.
UK v Secretary of State for the Home Department · UK Upper Tribunal (Immigration and Asylum Chamber)
“The Upper Tribunal pointedly noted that it cannot afford to have its limited resources absorbed by representatives who place false information before the Tribunal.”
briefonline.com.au ↗
The file →
  • UK v Secretary of State for the Home Department involved a solicitor on the record.
  • He filed grounds of appeal containing a false citation and initially denied AI use.
  • The tribunal noted the conduct but did not impose a sanction in this article.
  • UK Upper Tribunal (IAC), England and Wales.
  • March 23, 2026; the article says the solicitor likely used Google’s AI mode and had used ChatGPT for client emails and summaries.
Warned $1K ChatGPT 2026-03-20
Corwin St. Raymond was not fined, but Judge Carl Barbier issued him a formal warning over the AI-tainted filing in his department.
Gentry v. City of New Orleans · Corwin St. Raymond · U.S. District Court for the Eastern District of Louisiana · LA
wdsu.com ↗
The file →
  • Gentry v. City of New Orleans also involved Chief Deputy City Attorney Corwin St. Raymond.
  • The motion contained nine made-up case citations generated by AI.
  • Judge Carl Barbier issued a formal warning, not a fine.
  • The case was in the U.S. District Court for the Eastern District of Louisiana.
  • On March 20, 2026, the court warned him about improper AI use in his department.
Fined $500 ChatGPT 2026-03-19
A pro se plaintiff filed a Rule 11 sanctions motion containing fictitious and inaccurate AI-generated case citations; the court awarded fees to the defendant.
Brownfield v. Cherokee County School District No. 35 · Oscar Brownfield · United States District Court for the Eastern District of Oklahoma · OK
“Still, they “displayed a reckless disregard for the obligation to ensure that what is filed in federal court is true, verified, and worthy of the public trust,” warranting sanctions “both to redress the specific harm and to deter recurrence.””
fisherphillips.com ↗
The file →
  • Brownfield v. Cherokee County School District No. 35 involved a pro se plaintiff sanctioned for bogus citations.
  • His Rule 11 motion contained fictitious and inaccurate case citations generated by AI.
  • The court ordered $500 paid directly to the school district.
  • E.D. Okla., federal court in Oklahoma.
  • March 19, 2026; the amount reflected about two hours of defense counsel time.
Fined $2K ChatGPT 2026-03-10
A pro se plaintiff filed a brief with fabricated case citations and AI-generated content; the court imposed Rule 11 sanctions.
Allen v. Cass Casper · N.D. Ill. · IL
“Pro se status does not shelter plaintiffs from sanctions pursuant to Rule 11. … Filing a document that contains citations to nonexistent cases, quotes language that comes from no real case, or that contains arguments wholly unsupported by the record violates Rule 11. Allen did just that. This demonstrates that Allen failed to make a reasonable inquiry into the supporting law or facts.”
fisherphillips.com ↗
The file →
  • Allen v. Cass Casper involved a pro se plaintiff sanctioned over AI-made citations.
  • Her opposition brief contained at least two entirely fabricated case citations and AI-generated sections.
  • The court imposed a $1,500 Rule 11 monetary sanction.
  • N.D. Ill., federal court in Illinois.
  • March 10, 2026; the court said she relied too heavily on AI without reasonable inquiry.
Warned ChatGPT 2026-03-01
Several authorities and citations in judicial review grounds could not be located; the Tribunal found the errors consistent with AI-hallucination risk and accepted a supervising solicitor's explanation.
R(Munir) and SOSHD · Upper Tribunal Immigration and Asylum Chamber
““Freely available generative artificial intelligence tools, trained on a large language model such as ChatGPT are not capable of conducting reliable legal research.”
naturalandartificiallaw.com ↗
The file →
  • R(Munir) and SOSHD concerned false authorities and citations in judicial review grounds.
  • The Tribunal found the pattern of errors consistent with AI-hallucination risk.
  • The supervising solicitor accepted that false citations generated by AI may have been used.
  • Upper Tribunal Immigration and Asylum Chamber, UK.
  • 2026-03-01; the Tribunal stressed supervision and accurate checking of documents.
Fined $1K ChatGPT 2026-02-24
An attorney filed an error-filled brief drafted with ChatGPT and Westlaw Precision AI that cited numerous hallucinated cases, and Judge Brandon Long imposed a $1,000 sanction.
Woodward Harbor LLC v. City of Mandeville, et al. · Eastern District of Louisiana · LA
mblb.com ↗
The file →
  • Woodward Harbor LLC v. City of Mandeville, et al.; one attorney was sanctioned.
  • He submitted a brief drafted with ChatGPT and Westlaw Precision AI containing hallucinated cases.
  • The court imposed a $1,000 sanction.
  • Eastern District of Louisiana; federal; Louisiana.
  • 2026-02-24; the brief included at least 11 citations that were made up, misquoted, or misused.
Warned ChatGPT 2026-02-22
Three lawyers received a warning and two had to take a mandatory course after citing non-existent judgments generated by ChatGPT in court filings.
First disciplinary measures for lawyers · Judges in Arnhem, Rotterdam, and Groningen
verileges.nl ↗
The file →
  • Three lawyers were disciplined.
  • They cited non-existent judgments generated by ChatGPT.
  • The sanctions were a warning and mandatory course, with no amount stated.
  • Judges in Arnhem, Rotterdam, and Groningen.
  • 22 February 2026; the page says this was the first disciplinary measure.
Warned ChatGPT 2026-02-06
Kenosha County DA Xavier Solis filed a brief with undisclosed AI use and false legal citations; Judge David Hughes struck the filing and sanctioned him.
Xavier Solis brief incident · Xavier Solis · Kenosha County Circuit Court · WI
“Hughes struck Solis’ written response after finding it used AI without disclosure and relied on “hallucinated,” false legal citations.”
wjiinc.org ↗
The file →
  • Kenosha County DA Xavier Solis was sanctioned
  • He filed a response with undisclosed AI use and false legal citations
  • Judge David Hughes struck the filing and sanctioned him; no dollar amount was stated
  • Kenosha County Circuit Court in Wisconsin
  • Feb. 6 hearing; the article says the AI system was described as such as ChatGPT
Filing dismissed ChatGPT 2026-02-05
Defense counsel filed a motion-to-dismiss brief with false citations, then repeated the errors after a sanctions order to show cause; the court struck the filings and entered default judgment.
Flycatcher Corp. Ltd. v. Affable Ave. LLC · defendant’s counsel in Flycatcher · U.S. District Court for the Southern District of New York · NY
“Only 14 out of 60 cases were hallucinations.”

Cases that don't exist

  • 2022 WL 4637582
  • 2024 LEXIS 2847243
nysba.org ↗
The file →
  • Flycatcher Corp. Ltd. v. Affable Ave. LLC involved defendant’s counsel
  • Motion-to-dismiss papers contained 13 fake citations and eight real cases with fictitious quotes
  • Court struck the submissions and entered default judgment against the defendant
  • Southern District of New York, federal court
  • Opposing counsel suggested ChatGPT or another AI model; counsel kept filing flawed research
Fined $12K ChatGPT 2026-02-02
Lawyers filed briefs with ChatGPT-hallucinated citations and quotations, and Judge Julie A. Robinson publicly admonished and sanctioned them.
Lexos Media IP LLC v. Overstock.com Inc. · Sandeep Seth; Kenneth Kula; Christopher Joe; Michael Doell; David Cooper · D. Kan. · KS
“the Rule 11 duty is “nondelegable.””
minnlawyer.com ↗
The file →
  • Four lawyers for a plaintiff company were sanctioned in a patent infringement case.
  • They used ChatGPT to find caselaw and filed briefs with hallucinated citations.
  • Judge Robinson imposed $12,000 total in sanctions and publicly admonished them.
  • U.S. District Court for the District of Kansas, Kansas.
  • Order dated 2026-02-05; the court also revoked one lawyer’s pro hac vice admission and required self-reporting.
Fined $1K ChatGPT 2026-02-01
A New Orleans court fined John Walker $1,000 after his LSU Health Foundation brief contained at least 11 fabricated or mischaracterized citations.
John Walker · John Walker · New Orleans · LA
““To this day it blows my mind it has that capability,””
developmentcorporate.com ↗
The file →
  • John Walker was sanctioned over an LSU Health Foundation brief.
  • The brief contained at least 11 fabricated or mischaracterized citations.
  • The court imposed a $1,000 fine.
  • New Orleans, Louisiana state court.
  • February 2026; Walker used both ChatGPT and Westlaw Precision AI.
Warned ChatGPT 2026-01-29
The appellant used ChatGPT to draft legal submissions and cited cases the tribunal could not locate; the tribunal refused his application and noted possible referral if he were registered.
Folarin v The Immigration Services Commissioner [2026] UKFTT 135 · UK First-tier Tribunal
““This would have had the effect of misleading the Tribunal had the Tribunal not challenged the point.””
naturalandartificiallaw.com ↗
The file →
  • Folarin v The Immigration Services Commissioner; the appellant's submissions
  • He used ChatGPT and other AI tools to find and summarize cases, some of which could not be located
  • No formal sanction, but the tribunal found he was not fit to provide immigration advice
  • UK First-tier Tribunal
  • 2026; tribunal stressed that false citations could mislead and that a future regulator referral was possible
Fined $10K ChatGPT 2026-01-01
A lawyer filed a factum with non-existent authorities and was suspended for 6 months and fined $10,000.
Law Society of Ontario v Lee · Law Society of Ontario Tribunal
“To be clear, it is not professional misconduct for a lawyer to use AI to assist in the preparation of a factum.”
guides.library.queensu.ca ↗
The file →
  • Law Society of Ontario v Lee involved a lawyer sanctioned in 2026.
  • She filed a factum that included non-existent authorities and an authority that did not support her argument.
  • The LSO suspended her licence for 6 months and fined her $10,000.
  • The matter was before the Law Society of Ontario Tribunal in Canada.
  • The page says she used ChatGPT and quotes the tribunal on verifying AI-generated authorities.
Fined $3K ChatGPT 2025-12-12
Co-counsel Celeste Boyd used ChatGPT to draft and edit portions of briefs in an OnlyFans-related case and was sanctioned $3,000 for failing to verify them.
Celeste Boyd · Celeste Boyd · U.S. District Court for the Central District of California · CA
“The judge additionally imposed a $3,000 sanction on co-counsel Celeste Boyd, who he said used ChatGPT to draft and edit portions of the briefs but failed to verify the material, Reuters also reports.”
abajournal.com ↗
The file →
  • Celeste Boyd was sanctioned as co-counsel in the OnlyFans case.
  • She used ChatGPT to draft and edit portions of briefs that contained hallucinated material.
  • The court imposed a $3,000 sanction on Boyd.
  • U.S. District Court for the Central District of California; Judge Fred Slaughter; California.
  • The article says she failed to verify the material before filing.
Caught, not sanctioned ChatGPT 2025-12-10
A pro se claimant used ChatGPT to generate fabricated case law, and the tribunal awarded a preparation time order of £2,178 stayed pending appeal.
Mr J Harrison v Mr D May t/a Leeds Gymnastics Academy · Employment Tribunals (Leeds)
damiencharlotin.com ↗
The file →
  • Mr J Harrison v Mr D May t/a Leeds Gymnastics Academy involved AI-generated citations.
  • ChatGPT generated fabricated case law in the pleadings.
  • The tribunal awarded a preparation time order of £2,178, stayed pending EAT appeal.
  • Employment Tribunals (Leeds), UK.
  • 10 December 2025; the page says the payment was stayed pending appeal.
Fined $60K ChatGPT 2025-12-05
An Illinois circuit court sanctioned a law firm $50,000 and an attorney $10,000 for filing post-trial submissions with false AI-generated case citations and quotations.
Jordan v. Chi. Hous. Auth. · Ill. Cir. Ct. · IL
“The Court further emphasized its focus was “not on the misuse of artificial intelligence to conduct unreliable legal research and drafting,” but “the inexcusable submission of false authority and factual arguments to the Court, the subsequent misrepresentations about the extent of the improper conduct, and the failure to take prompt responsibility for errors once discovered.””

Cases that don't exist

  • Mack v. Anderson
klemchuk.com ↗
The file →
  • Jordan v. Chi. Hous. Auth.; Illinois law firm and attorney sanctioned
  • Post-trial filings contained false case citations and quotations generated by AI
  • $60,000 total sanctions: $50,000 firm and $10,000 attorney
  • Ill. Cir. Ct., Illinois
  • Dec. 5, 2025; drafting attorney said she used ChatGPT
Referred to the bar ChatGPT 2025-12-04
Jisuh Lee filed a factum with hallucinated case law and allegedly lied about using ChatGPT; the court referred the matter for criminal contempt proceedings.
Ko v. Li · Jisuh Lee of ML Lawyers · Ontario Superior Court of Justice
““This is a very unusual case and one that may resonate throughout the administration of justice.””
law360.ca ↗
The file →
  • Ko v. Li involved Jisuh Lee of ML Lawyers in an estate matter.
  • She filed a factum containing AI-hallucinated case law and later admitted using ChatGPT.
  • The court referred the contempt matter to Ontario’s attorney general; no amount was ordered.
  • Ontario Superior Court of Justice in Ontario, Canada.
  • Dec. 4 order; Justice Frederick Myers said it was a very unusual case.
Caught, not sanctioned ChatGPT 2025-12-04
McGuire Woods filed a brief with fake citations, and the court sanctioned the lawyer and ordered the firm to explain itself.
Ringer v. Bank of America · McGuire Woods LLP · U.S. District Court for the Northern District of Georgia · GA
“In other words, the Court must be concerned with the message sent to other lawyers, firms and non represented parties as to the consequences of cutting corners, by using tools such as ChatGPT or even AI modules offered by reputable legal research databases without adequate care, or otherwise failing to verify the accuracy of citations and assertions in legal filings.”
tcpaworld.com ↗
The file →
  • Ringer v. Bank of America; McGuire Woods LLP was involved.
  • A lawyer filed a brief containing a ton of fake cites.
  • The court sanctioned the lawyer and required the firm to explain itself.
  • U.S. District Court for the Northern District of Georgia.
  • Dec. 4, 2025; the court said careless citation errors must not be tolerated.
Referred to the bar ChatGPT 2025-11-27
A barrister used ChatGPT to draft appeal grounds with a false citation and the tribunal reported him to the Bar Standards Board but did not start contempt proceedings.
MS v Secretary of State for the Home Department · Upper Tribunal (Immigration and Asylum Chamber)
barcouncil.org.uk ↗
The file →
  • MS v Secretary of State for the Home Department involved a barrister before the Upper Tribunal.
  • He included a false citation and later used ChatGPT to draft the grounds of appeal.
  • The tribunal referred him to the Bar Standards Board but did not initiate contempt proceedings.
  • Court: Upper Tribunal (Immigration and Asylum Chamber), Bangladesh; [2025] UKUT 305 (IAC).
  • Date: 2025-11-27; he did not use Westlaw or Lexis Nexis to check the work.
Reprimanded ChatGPT 2025-11-21
Gordon Rees submitted filings with hallucinated AI citations; Judge Christopher L. Hawkins declined to sanction the firm but reprimanded former lawyer Cassie D. Preston.
Jackson Hospital & Clinic, Inc. bankruptcy proceeding · Gordon Rees Scully Mansukhani · U.S. Bankruptcy Court for the Middle District of Alabama · AL
“attorneys must not allow their personal feelings to cloud their professional judgment, and Ms. Preston crossed a line when she resorted to making use of arguments and authorities generated by artificial intelligence”
news.bloomberglaw.com ↗
The file →
  • Jackson Hosp. & Clinic Inc. and Cassie Preston were sanctioned in bankruptcy court.
  • She submitted filings with artificial-intelligence generated fake citations.
  • The court publicly reprimanded her and ordered disclosure/compliance steps, with no monetary amount stated.
  • US Bankruptcy Court for the Middle District of Alabama; Judge Christopher L. Hawkins.
  • Order dated 2025-11-21; the opinion also found misleading and fabricated citations.
Referred to the bar ChatGPT 2025-11-17
CLP filed judicial-review grounds with bogus authorities and the Upper Tribunal said the author and COLP should expect referral.
R (Munir) v SSHD · City Law Practice Solicitors and Advocates · Upper Tribunal (IAC)
“It would be easy to think that this is a case about the naïve use of generative AI, but it is not merely about that; it is principally about supervision and the obligation to ensure that the tribunal is not misled.”
asadakhan.wordpress.com ↗
The file →
  • R (Munir) v SSHD involved City Law Practice Solicitors and Advocates and Zubair Rasheed.
  • Judicial-review grounds were filed with bogus authorities and incorrect citations.
  • The Upper Tribunal said the conduct should ordinarily be referred to a regulator.
  • Upper Tribunal (IAC), United Kingdom; Judges Lindsley, Keith and Blundell.
  • 17 November 2025; the court stressed supervision failures and mentioned ChatGPT.
Filing dismissed ChatGPT 2025-11-14
A pro se filer used AI-generated citations in a bankruptcy filing; the court sanctioned her by dismissing the case without prejudice.
In re Cathleen Siebrecht · Cathleen Siebrecht · U.S. Bankruptcy Court, Eastern District of Virginia · VA
““The ‘AI-generated citations’ in the filing were ‘hallucinated’ and not supported by the law.””

Cases that don't exist

  • In re Zepeda, 2023 Bankr. LEXIS 1234
  • In re Jones, 2024 WL 1234567
calawyers.org ↗
The file →
  • In re Cathleen Siebrecht involved Cathleen Siebrecht, who filed a bankruptcy paper with AI-generated citations.
  • The filing contained hallucinated legal citations and the court found they were unsupported by law.
  • The court dismissed the case without prejudice.
  • It was in the U.S. Bankruptcy Court for the Eastern District of Virginia.
  • The order was entered on 2025-11-14, and the page says the citations were ChatGPT-generated.
Caught, not sanctioned ChatGPT 2025-11-10
Cal State filed a brief with faulty quotes and phantom quotations that an administrative law judge struck from the record.
California State University brief striking order · California State University · California Public Employment Relations Board · CA
“while “there is no proof that AI was, in fact the author” of a Cal State brief, the document “bears all the hallmarks of the hallucinations associated with AI-generated texts” and contains a series of misquotes Cal State failed to explain.”

Cases that don't exist

  • Marshall v. Regis Educational Corporation
laist.com ↗
The file →
  • California State University was the party whose brief was struck.
  • The brief contained faulty quotes and phantom quotations from a 1981 decision.
  • Administrative Law Judge Bernhard Rohrbacher ordered the filing struck from the record.
  • The matter was before the California Public Employment Relations Board in California.
  • CSU later acknowledged the brief had been written with AI assistance.
Referred to the bar ChatGPT 2025-10-29
Adam Hyman filed an AI-generated divorce brief with fake cases and misquotes, and the Appellate Court of Maryland referred him to the Attorney Grievance Commission.
Adam Hyman divorce brief matter · Adam Hyman · Appellate Court of Maryland · MD
“it is unquestionably improper for an attorney to submit a brief with fake cases generated by AI.”
thedailyrecord.com ↗
The file →
  • Adam Hyman was the lawyer referred to the Attorney Grievance Commission.
  • He submitted a divorce brief prepared with ChatGPT that contained fake cases and misquotes.
  • The court imposed a referral to the AGC, with no dollar amount stated.
  • The incident was in the Appellate Court of Maryland in Maryland.
  • The opinion was issued Oct. 29, 2025; Judge Kathryn Graeff said attorneys must verify cited authority.
Fined $1K ChatGPT 2025-10-23
Counsel filed opposition papers with AI-generated false Louisiana case citations; the trial court sanctioned her and the appellate court affirmed.
In re: Sanctions Order of KENNEY (2025) · Connie P. Trieu · Court of Appeal of Louisiana, Fifth Circuit · LA
“Plaintiff's counsel submitted legal authority and citations generated by “A.I.” software without verifying their authenticity.”

Cases that don't exist

  • Burns v. State, 813 So.2d 574
  • Smith v. Christus St. Patrick Hosp., 20-306 (La. App. 3 Cir. 11/3/20), 308 So.3d 1216
  • Doe v. Southwest Louisiana Hosp. Ass'n, 02-180 (La. App. 3 Cir. 7/10/02), 833 So.2d 1136
caselaw.findlaw.com ↗
The file →
  • In re: Sanctions Order of KENNEY (2025) sanctioned Connie P. Trieu.
  • She filed pleadings containing three fabricated Louisiana citations and other citation errors.
  • The trial court ordered $1,368 in attorney's fees plus costs.
  • Court of Appeal of Louisiana, Fifth Circuit, in Louisiana.
  • Decided October 23, 2025; the opinion says the citations were generated by AI and notes ChatGPT and Microsoft Copilot were used.
Referred to the bar ChatGPT 2025-10-06
A lawyer filed grounds of appeal with at least two non-existent authorities, and the judge ordered the solicitor to show cause why the matter should not be referred to the SRA.
AK v Secretary of State for the Home Department · Upper Tribunal
damiencharlotin.com ↗
The file →
  • AK v Secretary of State for the Home Department involved a lawyer's appeal grounds.
  • The false citations likely came from unchecked generative-AI drafting.
  • The judge directed the solicitor to show cause about an SRA referral.
  • Upper Tribunal, UK.
  • 6 October 2025; the page names ChatGPT as the AI tool.
Fined $5K ChatGPT 2025-10-01
Mr. Laprade filed proceedings with false AI-generated citations and other authorities, and the Superior Court of Quebec imposed a $5,000 penalty.
Specter Aviation Limited c. Laprade · Jean Laprade · Superior Court of Québec
“The Court therefore imposed a $5,000 sanction on the Defendant, both to punish his conduct and to deter similar behaviour in the future.”
gowlingwlg.com ↗
The file →
  • Specter Aviation Limited v. Laprade; Mr. Laprade was sanctioned
  • He filed proceedings with false citations, decisions, and references produced using AI
  • The court imposed a $5,000 penalty
  • Superior Court of Quebec, Quebec
  • October 1, 2025; the court said AI cannot justify inaccurate or fabricated documents
Caught, not sanctioned ChatGPT 2025-09-30
Jisuh Lee filed a factum with ChatGPT-generated fake cases and later admitted she misled the court about who prepared it; the court treated it as criminal contempt.
Ko v. Li · Jisuh Lee · Ontario Superior Court of Justice
““Delivering a factum containing non-existent legal submissions creates a serious risk of a miscarriage of justice.””
benmor.com ↗
The file →
  • Ko v. Li involved lawyer Jisuh Lee and a factum with fabricated authorities.
  • The article says she used ChatGPT and failed to verify the citations.
  • The court treated the matter as criminal contempt, with no sanction yet imposed on this page.
  • Ontario Superior Court of Justice in Ontario, Canada.
  • On 2025-09-30, Lee admitted she alone used ChatGPT and had lied about staff involvement.
Filing dismissed ChatGPT 2025-09-25
Oready, LLC filed protest briefs with fabricated, mismatched, and misquoted GAO citations, and GAO dismissed the protests as an abuse of process.
Oready, LLC · Government Accountability Office
“Our decision here is not based on the use of AI as a method of research; it is based on the protester’s repeated reliance on non-existent citations or decisions without verifying their correctness and applicability to the arguments being presented.”
centrelawgroup.com ↗
The file →
  • Oready, LLC was dismissed after filing AI-tainted protest briefs.
  • The filings included fabricated GAO decisions, mismatched citations, and misquoted holdings.
  • GAO dismissed the protests as an abuse of process.
  • Government Accountability Office; federal procurement protest forum.
  • Sept. 25, 2025; GAO said the issue was unverified non-existent citations, not AI use itself.
Fined $2K ChatGPT 2025-09-23
Lawrence Chang filed a reconsideration motion with AI-generated bogus citations and the court struck the motion and sanctioned him.
Lawrence Chang · Lawrence Chang · Texas state court · TX
“The court finds “good cause supports the imposition of sanctions in order to deter such conduct in the future, which amounts to serious misconduct that threatens to impugn the integrity of the courts and the administration of justice,” Guerra Gamble said.”
news.bloomberglaw.com ↗
The file →
  • Lawrence Chang was sanctioned in a Texas state-court matter involving his motion for reconsideration.
  • He admitted using AI and not checking the accuracy of the citations it produced.
  • The court ordered him to pay $2,000 in attorneys’ fees and expenses and struck the reconsideration motion.
  • Judge Maya Guerra Gamble issued the order in Texas state court.
  • Chang said he pasted his draft into ChatGPT’s legal tool to improve his writing.
Referred to the bar ChatGPT 2025-09-18
Barrister Muhammad Mujeebur Rahman cited a non-existent case generated with ChatGPT and was referred to the Bar Standards Board.
Y (China) · Muhammad Mujeebur Rahman · Upper Tribunal (Immigration and Asylum Chamber)
“We do however conclude that this is a case where referral to a regulator, in this instance the BSB, is most definitely appropriate”
localgovernmentlawyer.co.uk ↗
The file →
  • Muhammad Mujeebur Rahman was the barrister referred to the BSB.
  • He filed an immigration appeal relying on the bogus case Y (China) generated with ChatGPT.
  • The Upper Tribunal referred him to the Bar Standards Board; no contempt or police action was taken.
  • The court was the Upper Tribunal (Immigration and Asylum Chamber) in the UK.
  • On 2025-09-18, judges said he had not ensured the accuracy of what was placed before the tribunal.
Fined $10K ChatGPT 2025-09-12
Amir Mostafavi filed appellate briefs with fabricated AI-generated citations and quotations, and the California Court of Appeal sanctioned him $10,000 and referred him to the state bar.
Noland v. Land of the Free, L.P. · Amir Mostafavi · California Court of Appeal · CA
“Nearly all of the legal quotations in plaintiff’s opening brief, and many of the quotations in plaintiff’s reply brief, are fabricated,”

Cases that don't exist

  • Schimmel v. Levin (2011) 195 Cal.App.4th 81
counselpro.ai ↗
The file →
  • Noland v. Land of the Free, L.P.; attorney Amir Mostafavi was sanctioned.
  • He used ChatGPT and other AI tools to enhance appellate briefs without verifying citations.
  • The court imposed a $10,000 sanction and referred him to the state bar.
  • California Court of Appeal, California; the case was a state appellate decision.
  • The article says 21 of 23 opening-brief quotations were fabricated and the court faulted the filings.
Fined ChatGPT 2025-08-28
A pro se plaintiff filed a brief citing a non-existent Third Circuit case generated by AI, and Judge Karoline Mehalchick sanctioned him.
Allbaugh v. Univ. of Scranton · U.S. District Court for the Middle District of Pennsylvania · PA
“Rule 11 does not prohibit the use of generative AI, however, it does require parties to verify that they are not submitting briefs to the Court which rely on fictitious cases.”

Cases that don't exist

  • Allbaugh v. Univ. of Scranton, 2025 WL 2484188
cgalaw.com ↗
The file →
  • Allbaugh v. Univ. of Scranton involved a pro se plaintiff sanctioned for an AI-generated fake citation.
  • The plaintiff cited a non-existent Third Circuit opinion in a brief.
  • Judge Karoline Mehalchick imposed sanctions for the misuse of AI and the false citation.
  • The case was in the U.S. District Court for the Middle District of Pennsylvania.
  • On 2025-08-28, the court said parties must verify briefs do not rely on fictitious cases.
Fined ChatGPT 2025-08-14
In Mavy v. Commissioner of Social Security Administration, the court sanctioned Maren Ann-Miller Bam for a brief citing AI-generated fake cases.
Mavy v. Commissioner of Social Security Administration · Maren Ann-Miller Bam · U.S. District Court for the District of Arizona · AZ
“The Court does not take this action lightly.”

Cases that don't exist

  • Brown v. Colvin
  • Wofford v. Berryhill
  • Hobbs v. Commissioner of Social Security Administration
cronkitenews.azpbs.org ↗
The file →
  • Mavy v. Commissioner of Social Security Administration; attorney Maren Ann-Miller Bam was sanctioned.
  • Her Social Security appeal brief included AI-generated fabricated, misleading, or unsupported citations.
  • Judge Alison Bachus sanctioned her and ordered notice to the three judges named in the fake opinions.
  • U.S. District Court for the District of Arizona; federal court in Arizona.
  • Aug. 14, 2025; the article says 12 of 19 cited cases were fake and mentions ChatGPT-generated hallucinations.
Referred to the bar ChatGPT 2025-08-12
A barrister filed ChatGPT-drafted grounds with a fictitious Court of Appeal authority and was found to have attempted to mislead the tribunal and referred to the BSB.
MS (Bangladesh) · Upper Tribunal (Immigration and Asylum Chamber)
“attempted to mislead”
searchthe.law ↗
The file →
  • MS (professional conduct: AI generated documents) is the named matter.
  • ChatGPT-drafted grounds cited a fictitious Court of Appeal authority.
  • The barrister was referred to the Bar Standards Board.
  • The court was the Upper Tribunal (IAC).
  • Promulgated 12 August 2025; the barrister produced further unverified AI output after a short adjournment.
Referred to the bar $100K ChatGPT 2025-07-24
A federal judge sanctioned three Butler Snow LLP attorneys for filing motions written by generative AI that included false case citations.
Frankie Johnson v. Jefferson S. Dunn et al. · Butler Snow attorneys · United States District Court for the Northern District of Alabama · AL
“"failing to prevent AI hallucinations in legal filings"”
prnewswire.com ↗
The file →
  • Johnson v. Dunn; three Butler Snow LLP attorneys
  • Filed five fabricated citations in a case
  • Disqualified and referred to bar regulators in every state where they were licensed
  • U.S. District Court for the Northern District of Alabama
  • July 23, 2025; Judge Anna M. Manasco issued the order
Filing dismissed $86K ChatGPT 2025-07-17
Lawyer James Martin Paul filed briefs with hallucinated AI citations in ByoPlanet cases; the court dismissed the federal matters, awarded fees, and referred him to the Florida Bar.
ByoPlanet International, LLC v. Johansson · James Martin Paul · Southern District of Florida · FL
“When he argued the fine was too high, the court shot back that caving to his arguments “would only benefit serial hallucinators.””
relativity.com ↗
The file →
  • ByoPlanet Int’l, LLC v. Johansson and related matters; attorney James Martin Paul
  • Paul used hallucinated citations across eight matters and even filed fabricated quotations after being put on notice
  • Judge Leibowitz dismissed the four federal matters without prejudice and without leave to amend, awarded fees and costs, and referred Paul to the Florida Bar
  • U.S. District Court for the Southern District of Florida, Florida
  • July 17, 2025; Paul told the court he and his paralegal used ChatGPT
Referred to the bar ChatGPT 2025-06-06
The claimant's solicitor relied on 45 client-provided case citations, 18 fictitious, and the court referred him to the SRA.
Hamad Al Haroun v Qatar National Bank [2025] EWHC 1383 (Admin) · Abid Hussain; Primus Solicitors · Divisional Court
“a matter of the utmost seriousness”
dacbeachcroft.com ↗
The file →
  • Hamad Al-Haroun v Qatar National Bank QPSC and QNB Capital LLC; Abid Hussain of Primus Solicitors
  • Witness statements contained eighteen non-existent case authorities and misquoted genuine ones
  • The court did not initiate contempt proceedings and said referral to the regulator was more appropriate
  • Divisional Court, UK; the article says the decision was published on 6 June 2025
  • Dias J called it "a matter of the utmost seriousness" and said counsel failed to check the information provided by the client
Caught, not sanctioned ChatGPT 2025-05-20
Senior lawyer Jisuh Lee showed cause over fake precedent cases in a factum, but she apologized and purged her contempt before the hearing continued.
Ko v Li · Jisuh Lee · Ontario Superior Court of Justice
“to bring home to all lawyers the need to check and not to trust factums [sic] generated by AI or by others”
claytonrice.com ↗
The file →
  • Ko v Li involved senior lawyer Jisuh Lee.
  • Her factum contained fake precedent court cases prepared using ChatGPT.
  • She purged her contempt, so no sanction was imposed.
  • Ontario Superior Court of Justice.
  • 2025-05-20; Justice F.L. Myers said lawyers must check factums generated by AI.
Filing dismissed ChatGPT 2025-04-01
An applicant repeatedly used the term "subornation of perjury," which Judge David Nolan said sounded like AI and denied the judicial review appeal.
Judicial review appeal in Ireland · High Court
“This sounds like something that derived from an artificial intelligence source,”
businessplus.ie ↗
The file →
  • An unnamed applicant was before Judge David Nolan.
  • The applicant repeatedly used a phrase the judge thought came from AI.
  • Judge Nolan denied the judicial review appeal.
  • This happened in the High Court in Ireland.
  • About four months after Dec. 2024; Nolan said it had the hallmarks of ChatGPT or a similar tool.
Referred to the bar ChatGPT 2025-03-31
The court found seven of nine cited authorities were non-existent, apparently AI-generated, refused leave to appeal, and referred the matter to the Legal Practice Council.
Mavundla v MEC: Department of Co-Operative Government and Traditional Affairs KwaZulu-Natal and Others [2025] ZAKZPHC 2 · Pietermaritzburg High Court · South Africa
“irresponsible and downright unprofessional”
iol.co.za ↗
The file →
  • Mavundla v MEC: Department of Co-Operative Government and Traditional Affairs KwaZulu-Natal and Others
  • Seven of nine authorities in the applicant’s submissions were found to be entirely non-existent, apparently generated by AI.
  • The court refused leave to appeal and referred the matter to the Legal Practice Council.
  • Pietermaritzburg High Court, South Africa.
  • Judge Bezuidenhout; the article says this was the clearest African example of AI hallucinations contaminating filings.
Fined $2K ChatGPT 2025-02-27
Counsel admitted using ChatGPT to draft filings with fabricated and misrepresented citations; the court imposed a $2,500 sanction and mandatory AI-ethics CLE.
Bunce v. Visual Tech. Innovations, Inc. · U.S. District Court for the Eastern District of Pennsylvania · PA
damiencharlotin.com ↗
The file →
  • Bunce v. Visual Technology Innovations (1); counsel sanctioned.
  • The page says counsel admitted using ChatGPT to draft filings with fake citations.
  • The court imposed a $2,500 monetary sanction and required CLE.
  • E.D. Pennsylvania, federal court in Pennsylvania.
  • 2025-02-27; the matter involved fabricated, misrepresented, and outdated citations.
Referred to the bar ChatGPT 2025-01-31
A lawyer filed submissions with false citations and quotes generated by ChatGPT, and Judge Skaros referred the conduct to the regulatory body.
Valu v Minister for Immigration and Multicultural Affairs (No 2) · Federal Circuit and Family Court of Australia (Division 2)
“The conduct of the ALR, in filing an application and submissions which contained citations to Federal Court of Australia cases which do not exist and alleged quotes from the Tribunal’s decision which do not exist, falls short of the standard of competence and diligence that the applicant in the substantive proceedings was entitled to expect from his legal representative.”
lst.org.au ↗
The file →
  • Valu v Minister for Immigration and Multicultural Affairs (No 2) involved a lawyer referred to the regulatory body.
  • The filings contained false citations and quotes generated using ChatGPT.
  • Judge Skaros referred the lawyer's conduct for consideration.
  • Federal Circuit and Family Court of Australia (Division 2).
  • 2025-01-31; the conduct also involved correspondence with the Court without the consent of the other parties.
Warned ChatGPT 2025-01-10
Counsel submitted an expert declaration drafted with LLM help that cited two nonexistent articles and misattributed a third; the court excluded the testimony.
Kohls v. Ellison · Jeff Hancock · United States District Court for the District of Minnesota · MN
“Rule 11 imposes a “personal, nondelegable responsibility” to ensure the accuracy of everything filed.”
counselpro.ai ↗
The file →
  • Kohls v. Ellison; Minnesota Attorney General Ellison’s counsel
  • An expert declaration used LLM help and contained two nonexistent articles and one misattribution
  • The court excluded the expert testimony; no monetary sanction stated
  • U.S. District Court for the District of Minnesota, Minnesota
  • Jan. 10, 2025; the court stressed Rule 11’s personal, nondelegable responsibility
Filing dismissed ChatGPT 2025-01-01
The court dismissed a damages claim after the plaintiff's ChatGPT-assisted complaint relied on inaccurate or non-existent legal bases.
AI-written lawsuit dismissed by the court · Sąd Okręgowy we Wrocławiu
“the statement of claim having been drafted entirely with the assistance of ChatGPT.”
konbriefing.com ↗
The file →
  • AI-written lawsuit dismissed by the court; the plaintiff
  • The complaint was drafted entirely with ChatGPT and used inaccurate or non-existent legal bases
  • The court dismissed the claim as unfounded
  • Wrocław, Poland
  • 2025; the court said responsibility for the content rested with the plaintiff
Caught, not sanctioned ChatGPT 2025-01-01
The appellant filed written submissions with six inaccurate cited decisions and one apparently fictitious case; the court ignored the authorities and denied costs.
NCR v KKB, 2025 ABKB 417 · Alberta Court of King’s Bench
“Citing fake cases in court filings and other materials handed up to the court is an abuse of process and is tantamount to making a false statement to the court. Unchecked, it can lead to the miscarriage of justice.”
hayherlee.com ↗
The file →
  • NCR v KKB, 2025 ABKB 417 involved a self-represented mother.
  • Her written submissions cited six inaccurate decisions and one apparently fictitious case.
  • The court denied costs and ordered each party to bear their own costs.
  • The decision was from the Alberta Court of King’s Bench.
  • The court inferred possible AI-generated content and quoted Zhang v Chen about fake cases.
Referred to the bar ChatGPT 2025-01-01
A barrister cited a false case generated by ChatGPT and the Upper Tribunal referred the barrister to the BSB after the citation was not checked.
MS (Bangladesh) (Professional Conduct: AI Generated Documents) · Upper Tribunal
5essex.co.uk ↗
The file →
  • MS (Bangladesh) (Professional Conduct: AI Generated Documents) involved a barrister.
  • The barrister cited a false case generated by ChatGPT and failed to check its authenticity.
  • The Upper Tribunal referred the barrister to the BSB.
  • Upper Tribunal, location not stated on the page.
  • The article says the tribunal applied Ayinde guidance and the false citation came from ChatGPT.
Caught, not sanctioned ChatGPT 2024-12-01
A tribunal order was reportedly withdrawn after it cited a wrong High Court, a wrong party name, and a repealed provision.
Buckeye Trust v. PCIT · ITAT Bangalore · KA
advocateprakhar.com ↗
The file →
  • Buckeye Trust v. PCIT was identified as the first documented Indian AI-hallucination incident.
  • The ITAT Bangalore order was reportedly withdrawn after suspected AI hallucinations.
  • The issue was wrong citations, including a wrong High Court and wrong party name.
  • The outcome was withdrawal and re-hearing, not a monetary penalty.
  • The page dates it to December 2024 and says the tribunal order was later identified retrospectively.
Caught, not sanctioned ChatGPT 2024-12-01
The Bengaluru bench recalled an order after it cited four non-existent judgments traced to unverified ChatGPT use.
Income Tax Appellate Tribunal (ITAT) Bengaluru bench matter · Income Tax Appellate Tribunal, Bengaluru bench
medianama.com ↗
The file →
  • ITAT Bengaluru bench matter
  • An order cited four non-existent judgments from unverified ChatGPT use
  • The bench recalled the order; no formal sanction is stated
  • Income Tax Appellate Tribunal, Bengaluru
  • December 2024; the citations were traced back to ChatGPT
Fined $2K ChatGPT 2024-11-04
Judge Fred W. Slaughter imposed $2,000 in sanctions on William Becker Jr. for filing a motion with alleged non-existent case citations and misrepresentations of existing law.
William Becker Jr. · William Becker Jr · US District Court for the Central District of California · CA
“Sanctions were warranted against Becker for “his failure to verify the validity of the AI-generated material, and submitting the AI-tainted Motion to the court,” the judge said.”
news.bloomberglaw.com ↗
The file →
  • William Becker Jr. was sanctioned in the Chris Kluwe-related case.
  • He filed a motion with AI-hallucinated/non-existent citations and misrepresented existing case law.
  • The court imposed $2,000 in sanctions.
  • US District Court for the Central District of California; Judge Fred W. Slaughter.
  • Nov. 4 order; the page says the motion was AI-tainted and Becker was told to show cause.
Fined $2K ChatGPT 2024-02-12
A brief citing fictitious cases created with AI led the Superior Court of Massachusetts to sanction the attorney $2,000.
Smith v. Falwell · Suffolk Superior Court · MA
“two disturbing developments that are adversely affecting the practice of law”
klemchuk.com ↗
The file →
  • Smith v. Falwell involved lead counsel at the offending law firm.
  • Three filings contained fictitious or non-existent caselaw produced using AI.
  • The court sanctioned the attorney $2,000.
  • Suffolk Superior Court in Massachusetts.
  • Feb. 12, 2024; the article says ChatGPT and Google Bard were cited as examples of AI that can fabricate information.
Referred to the bar ChatGPT 2024-01-30
The Second Circuit referred attorney Jae S. Lee to its grievance panel after he cited a ChatGPT-generated non-existent case in a reply brief.
Park v. Kim · Jae S. Lee · U.S. Court of Appeals for the Second Circuit · NY
“well below the basic obligations of counsel”
getvoibe.com ↗
The file →
  • Park v. Kim involved Jae S. Lee of JSL Law Offices.
  • She cited a non-existent case in a reply brief and admitted ChatGPT use.
  • The Second Circuit referred her to its Grievance Panel.
  • U.S. Court of Appeals for the Second Circuit.
  • January 30, 2024; the panel said her conduct was well below the basic obligations of counsel.
Filing dismissed $10K ChatGPT 2024-01-01
A litigant submitted an appellate brief with fictitious AI-generated case citations, and the Missouri Court of Appeals dismissed the appeal and imposed damages.
Kruse v. Karlen · Jonathan Karlen · Missouri Court of Appeals · MO
“The Court of Appeals noted that the “bogus citations” could not, “for any reason….be countenanced and represent a flagrant violation of the duty of candor” owed to the court.”
nysba.org ↗
The file →
  • Kruse v. Karlen involved a self-represented appellant.
  • Her AI program cited 24 cases, but only two were real and both were misunderstood.
  • The court dismissed the appeal and ordered $10,000 in attorney’s fees.
  • Missouri Court of Appeals, Missouri.
  • The article says the AI made up case names and citations that appeared genuine.
Fined ChatGPT 2024-01-01
A lawyer filed a brief quoting two made-up cases, and the court ordered costs and required a review of all files for ChatGPT use.
Zhang v. Chen · British Columbia Supreme Court
“As this case has unfortunately made clear, generative AI is still no substitute for the professional expertise that the justice system requires of lawyers.”
calgarydefence.com ↗
The file →
  • Zhang v. Chen involved the lawyer representing Mr. Chen.
  • The brief quoted two made-up cases and used ChatGPT.
  • The court ordered costs and a file-wide review, not a stated dollar fine.
  • British Columbia Supreme Court, British Columbia.
  • 2024 BCSC 285; Justice Masuhara said generative AI is no substitute for lawyers' expertise.
Caught, not sanctioned ChatGPT 2023-11-22
Attorney Zachariah Crabill received a 90-day suspension after filing fabricated ChatGPT citations in a custody case and lying to the judge about their origin.
People v. Crabill · Zachariah Crabill · Colorado Office of Presiding Disciplinary Judge · CO
“In November 2023, a Colorado disciplinary judge approved a suspension of one year and one day, with 90 days to serve, for violations including competence, diligence, and candor to the tribunal.”
getvoibe.com ↗
The file →
  • Zachariah Crabill was sanctioned in a Colorado custody case.
  • He filed fabricated ChatGPT citations and then lied to the judge about their origin.
  • The court imposed a 90-day suspension.
  • Colorado state court.
  • 2024; the page says this was not merely a fine.
Warned ChatGPT 2023-08-22
The Delhi High Court warned that ChatGPT output cannot be the basis of adjudication and cautioned against fictional case laws and imaginative data.
Christian Louboutin SAS v. M/s The Shoe Boutique-Shutiq · Delhi High Court · DL
“the responses generated by ChatGPT cannot be the basis of adjudication of legal or factual issues in a court of law.”
blog.ipleaders.in ↗
The file →
  • Christian Louboutin SAS v. M/s The Shoe Boutique-Shutiq
  • A party had leaned on ChatGPT in a trademark dispute
  • The court issued a warning, not a money sanction
  • Delhi High Court, Delhi
  • 22 Aug 2023; the court warned against fictional case laws
Fined $10K ChatGPT 2023-06-30
Morris, Nichols, Arsht & Tunnell LLP submitted a brief with hallucinated citations, and the court sanctioned the firm and counsel.
In re: CFI Delegated Shareholder Litigation · Morris, Nichols, Arsht & Tunnell LLP · Court of Chancery of the State of Delaware · DE
““The use of ChatGPT was not disclosed to the Court.””

Cases that don't exist

  • Hsu v. Teco Energy, Inc., 2018 WL 5115675
  • Morris v. 7-Eleven, Inc., 2015 WL 7293045
podcasts.apple.com ↗
The file →
  • In re: CFI Delegated Shareholder Litigation against Morris, Nichols, Arsht & Tunnell LLP
  • A filing included hallucinated citations generated with ChatGPT
  • The court imposed a $10,000 monetary sanction
  • Court of Chancery of the State of Delaware, Delaware
  • The order noted the use of ChatGPT was not disclosed to the court
Warned ChatGPT 2023-06-22
Levidow, Levidow & Oberman P.C. filed a submission with fake citations, and the court warned of sanctions but did not impose one.
Moten v. Apfel · Levidow, Levidow & Oberman P.C · U.S. District Court for the Southern District of New York · NY
““The Court is deeply concerned that a lawyer would submit a brief with citations that do not exist.””

Cases that don't exist

  • Martinez v. Air Line Pilots Ass’n, 2021 WL 4314517
  • Moten v. Apfel, 2023 WL 4001202
podcasts.apple.com ↗
The file →
  • Moten v. Apfel against Levidow, Levidow & Oberman P.C.
  • A filing contained AI-fabricated legal citations
  • The court issued a warning rather than a sanction
  • U.S. District Court for the Southern District of New York, New York
  • The judge said the court was deeply concerned about nonexistent citations
Fined $5K ChatGPT 2023-06-22
Steven Schwartz filed six ChatGPT-invented cases in a brief, and Judge P. Kevin Castel sanctioned him, co-counsel Peter LoDuca, and their firm.
Mata v. Avianca, Inc. · Steven Schwartz, Peter LoDuca, and their firm · U.S. District Court for the Southern District of New York · NY
“Judge P. Kevin Castel sanctioned Schwartz, co-counsel Peter LoDuca, and their firm $5,000.”

Cases that don't exist

  • Miller v. DLA Piper LLP (US), 923 F.3d 1297
  • Zicherman v. Korean Air Lines Co., 516 U.S. 217
  • Varghese v. China Southern Airlines Co. Ltd., 925 F.3d 1339

+3 more

gc.ai ↗
The file →
  • Mata v. Avianca; Steven Schwartz, Peter LoDuca, and their firm
  • Schwartz used ChatGPT and filed six nonexistent decisions with fabricated quotes and case numbers
  • Monetary fine: $5,000
  • U.S. District Court for the Southern District of New York
  • June 22, 2023; Judge P. Kevin Castel found they stood behind the fake cases after warning
Fined ChatGPT 2023-06-22
A New York lawyer was sanctioned after filing briefs with fake court cases generated by ChatGPT.
In re Cathleen Siebrecht · Cathleen Siebrecht · U.S. District Court for the Southern District of New York · NY
smallwarsjournal.com ↗
The file →
  • In re Cathleen Siebrecht involved Cathleen Siebrecht, a New York lawyer.
  • She filed briefs citing fake court cases produced by ChatGPT.
  • The court sanctioned her, but the article gives no amount.
  • U.S. District Court for the Southern District of New York.
  • The page says this happened in 2023 and ties the citations to a ChatGPT hallucination.
Warned ChatGPT 2023-05-27
A lawyer submitted a filing with six non-existent cases invented by ChatGPT, and the judge called it absurd and considered sanctions.
A lawyer submitted a filing with six non-existent cases invented by ChatGPT · New York judge · NY
“The judge called it absurd and considered sanctions.”
verileges.nl ↗
The file →
  • A lawyer filed a New York submission with fabricated cases.
  • ChatGPT invented six non-existent cases in the filing.
  • The judge considered sanctions but the page does not say they were imposed.
  • New York.
  • 27 May 2023; the judge called it absurd.
Caught, not sanctioned ChatGPT 2023-01-01
A litigant in person filed an appeal submission citing nine non-genuine authorities that were hallucinated by generative AI.
Harber v Commissioners for His Majesty’s Revenue and Customs · UKFTT
hausfeld.com ↗
The file →
  • Harber v Commissioners for His Majesty’s Revenue and Customs involved a litigant in person
  • She cited nine supportive authorities in an appeal submission
  • The authorities were fabricated and hallucinated by generative AI
  • No sanction is reported on this page
  • UKFTT; the page gives 2023 and identifies ChatGPT as an example
Fined $15K ChatGPT 2023-01-01
Two lawyers filed a brief with fake cases after one used ChatGPT; the court imposed a $5,000 fine on each lawyer and the firm.
US 2023 sanctions matter involving two lawyers and their firm
“The summaries of the decisions cited in the brief showed “stylistic and reasoning flaws that do not generally appear in decisions issued by United States Courts of Appeals.””
posteo.de ↗
The file →
  • Two lawyers and their law firm were sanctioned in the US.
  • They filed a brief citing non-existent cases and could not produce them.
  • The court imposed a $5,000 fine on each lawyer and the firm, totaling $15,000.
  • US court; federal jurisdiction implied by the cited courts of appeals.
  • In 2023; one lawyer had asked ChatGPT to provide supporting evidence.
Caught, not sanctioned ChatGPT 2023-01-01
The page says this 2023 South African case dealt with fake case law produced by ChatGPT, and the judge was more forgiving and found no intent to mislead.
Parker v Forsyth
theconversation.com ↗
The file →
  • Parker v Forsyth is named as a 2023 South African case.
  • The page says fake case law was produced by ChatGPT.
  • The judge found no intent to mislead and was more forgiving.
  • The court and judge are not named on the page.
  • 2023; the page presents it as an earlier, less severe South African incident.
Caught, not sanctioned ChatGPT 2023-01-01
An appellant filed nine bogus historical tribunal decisions as precedents and said it was possible she had used ChatGPT.
UK tax tribunal case · UK tax tribunal
lawsociety.org.zw ↗
The file →
  • UK tax tribunal case involving an appellant
  • Nine bogus historical tribunal decisions were submitted as supposed precedents
  • The article reports no sanction; the appellant said it was possible she used ChatGPT
  • UK tax tribunal
  • 2023; the page says this was one of the earlier AI-hallucination cases
Fined $5K ChatGPT 2023-01-01
A lawyer was challenged over seven apparently fictitious cases, asked ChatGPT to summarize them, and the court imposed a $5,000 fine on the two lawyers and their firm.
2023 case in the US district court for the southern district of New York · US District Court for the Southern District of New York · NY
“the result, said the judge, was “gibberish””
lawsociety.org.zw ↗
The file →
  • 2023 SDNY case involving a lawyer, two lawyers, and their firm
  • Seven apparently fictitious cases were cited and ChatGPT was used to summarize them
  • The court fined the two lawyers and their firm $5,000
  • US District Court for the Southern District of New York
  • 2023; the judge described the ChatGPT output as “gibberish”
Get Sanctions Watch in your inbox
New AI-citation sanctions, the morning after they hit the docket.