October 18, 2026 (in 16 days): Tenth Circuit: comments close on 2027 rules, incl. proposed Rule 46.5(B)(4) generative-AI review →

WRI Summit REIT LP v. Kuerschner

Arizona Court of Appeals, Division One · Ariz. Ct. App. · Arizona bar guidance

Pro-se party

Conduct

Self-represented guarantor filed briefs with no record citations, misquoted cases, and one apparently nonexistent case, despite a prior warning; the reply followed more.

Consequence

Appeal dismissed under ARCAP 25; opponent awarded appellate fees for the ARCAP 13 violations; a months-late corrective motion was denied.

Lesson

Moving to withdraw incorrect case descriptions months after briefing closed was, in the court's words, 'too little, too late.'

Court sanction

Verified September 28, 2026

Citation
WRI Summit REIT LP v. Kuerschner, No. 1 CA-CV 25-0854 (Ariz. App. Aug. 10, 2026) (mem. decision)
Decided
August 10, 2026

Summary

Self-represented appellant Stefanie Kuerschner appealed judgments awarding WRI Summit REIT, LP $146,000 in unpaid rent on its lease to Summit Aesthetics, LLC, whose performance she had guaranteed, and sanctioning her for improper subpoenas served on the personal representative of her late husband's estate. Her opening brief did "not include a single citation to the record," and it "consistently misquotes cases and cites cases for propositions either not mentioned or directly contradicted by the case. And one case she cites appears not to exist." The court gave examples: language attributed to State ex rel. Dep't of Econ. Sec. v. Burton and Bennett v. Baxter Grp. that does not appear in either; a due-process quotation attributed to Brown v. U.S. Fidelity & Guar. Co., which "did not mention due process at all"; a Cruz v. Superior Court citation the court could not locate; and a description of Takieh v. O'Meara as reversing a fee sanction when it affirmed one. The court stated: "These citation errors are consistent with the misuse of generative AI." Her reply brief, filed after further warnings, again lacked record citations and miscited authority, including presenting a 2014 memorandum decision as a published opinion.

AI tool:
Suspected generative AI (the court stated the errors are "consistent with the misuse of generative AI"; no finding, admission, or tool)
Sanction amount:
Appeal dismissed under ARCAP 25; personal representative awarded reasonable appellate attorney fees and taxable costs under ARCAP 25 (amount to be set under ARCAP 21); late corrective motion denied
This case summary is informational only. Verify the underlying opinion or order against the primary source before relying on it in any filing or client matter.

What sanction did the court impose?

In a memorandum decision by Presiding Judge Perkins (joined by Judges Furuya and Gaona), the court dismissed the appeal under ARCAP 25. It relied on her repeated notice: a September 2025 decision in her probate appeal that had flagged missing record citations and inaccurate case citations, a January 2026 order in this appeal warning of sanctions for "miscited, misrepresented, or nonexistent legal authorities," and a March 2026 decision dismissing another of her appeals for "repeated and unrepentant violations" of ARCAP 13. It denied her July 16 "Corrective Motion to Rely on Verified Authorities" as "too little, too late" because it came several months after briefing closed. It awarded the personal representative his appellate fees and costs under ARCAP 25 for her ARCAP 13 violations, and awarded WRI Summit its fees and costs under the guaranty's indemnity provision.

Why does WRI Summit REIT LP v. Kuerschner matter for law firms using AI?

Division One treated dismissal as a considered choice rather than an automatic consequence. It observed that it could dismiss for ARCAP 13 violations under Acciavatti, but that “we generally prefer to resolve cases on their merits,” and so examined “the circumstances of [this] case.” What tipped the balance was notice. The court traced three prior warnings: its September 2025 decision in her probate appeal, a January 2026 order in this appeal about “miscited, misrepresented, or nonexistent legal authorities,” and a March 2026 decision dismissing another of her appeals. Her reply brief, filed after all three, repeated the same kinds of errors.

The court’s AI language is measured. It did not find that she used generative AI; it observed that the pattern of misquotation, misattribution, and a case it could not locate was “consistent with the misuse of generative AI.” The operative violations were ARCAP 13’s requirements of record citations and accurate legal authority, which do not depend on how the errors arose.

The treatment of the late correction is the most practical point. The court said it appreciated her candor and “always encourage[s] litigants to correct mistakes in their briefs,” but a motion filed “several months after the briefing was completed” was “too little, too late.” Firms documenting compliance may wish to consider that the timing of a correction, not only its existence, can determine whether it mitigates the consequence.

Implications for your firm

Operational steps a firm reading this case may wish to consider documenting. Strategic and rule-application calls belong to your firm's attorneys.

  • Document a correction protocol that acts promptly. The appellant eventually moved to withdraw the incorrect case descriptions, but the court denied the motion because it came months after briefing ended.
  • Review a repeat litigant's prior appellate decisions for citation warnings; the court relied on warnings in two earlier appeals and an order in this one to justify dismissal.
  • Consider an ARCAP 25 fee request where an opponent's briefing violates ARCAP 13. The personal representative requested fees under A.R.S. 12-349(A)(1)-(3) and ARCAP 25; the court awarded them under ARCAP 25 because she failed to cite the record and made many inaccurate and misleading legal citations.

Sources

Primary sources

Unverified claims:
  • AI attribution: the decision says only that the citation errors are 'consistent with the misuse of generative AI'; it makes no finding that she used AI and she did not admit it. The attribution beyond that statement is a Charlotin tracker inference ('Implied').