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

Maxwell v. Michael

U.S. District Court, Southern District of Indiana, Indianapolis Division · S.D. Ind. · Indiana bar guidance

Pro-se party

Conduct

A pro se plaintiff filed briefs the defendants showed contained 28 quotations absent from the cited cases and nine citations to cases that do not appear to exist.

Consequence

Here the magistrate judge imposed a $1,000 Rule 11 sanction, reduced from the $9,250 the defendants sought, citing deterrence and the plaintiff's indigence.

Lesson

Here the court held that formatting 'placeholder' citations as real ones showed intent to deceive, whether or not AI was involved.

Court sanction

Verified September 28, 2026

Citation
Maxwell v. Michael, No. 1:24-cv-01365-JRS-TAB, Order on Plaintiff's Show Cause Response and Sanctions (S.D. Ind. May 26, 2026) (Baker, M.J.), ECF No. 206
Decided
May 26, 2026

Summary

William C. Maxwell, proceeding pro se, sued two Indiana State Police officers under Section 1983 and Indiana law. The defendants moved for an order to show cause, identifying 28 quotations that do not appear in the cited cases and nine citations to cases that do not appear to exist, 37 improper citations in all. On March 26, 2026, the district judge granted that motion and designated Magistrate Judge Tim A. Baker to decide whether Rule 11(c) sanctions were warranted. Maxwell's response did not dispute that certain citations were incorrect; he said his errors were not intentional, that some citations were "placeholders" he "intended to find support for" and correct before filing, and that some quotations were paraphrases. The court noted, as the defendants had, that his show-cause response did not confirm whether he used generative AI, agreed with the defendants that the fabricated authority "contains the hallmarks of generative AI hallucinations," and held that "[w]hether or not Plaintiff used AI, his conduct remains sanctionable, as it represents a deliberate intent to fabricate citations and quotations."

AI tool:
Suspected, not admitted (the court agreed with the defendants that the fabricated authority "contains the hallmarks of generative AI hallucinations"; the plaintiff's response did not confirm AI use)
Sanction amount:
$1,000 Rule 11 sanction payable to the Clerk by June 25, 2026 (plus a separate $886.80 in deposition costs owed to the defendants on an earlier discovery sanction)
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?

Magistrate Judge Baker sanctioned Maxwell $1,000 under Rule 11, payable to the Clerk of Court by June 25, 2026. The defendants had asked for $250 per fake citation or quotation, $9,250 in total; the court reduced the amount, citing Rule 11(c)(4)'s deterrence limit and Maxwell's indigent status, and described the figure as comparable to or less than sanctions recently imposed on self-represented litigants in the Seventh Circuit for similar Rule 11 violations. The court reasoned that fabricating a case name, number, citation, pin cite, and court and formatting it as a legitimate citation "shows an intent to deceive the Court, not simply a reckless failure to verify," and noted that Maxwell made no attempt to explain why he presented paraphrases as direct quotations. In the same order the court fixed the deposition costs owed under an earlier, separate discovery sanction at $886.80, payable to the defendants by the same date. The order notes that two motions to reconsider the March 26, 2026 order remained pending before the district judge.

Why does Maxwell v. Michael matter for law firms using AI?

Maxwell turns on intent rather than on AI. The court recounts the defendants’ argument that Maxwell acted in bad faith or in reckless reliance on AI, and it agreed that the fabricated authority bore “the hallmarks of generative AI hallucinations,” citing the Seventh Circuit’s discussion of pro se AI use in Jones v. Kankakee County Sheriff’s Department. But it did not rest the sanction on AI use. Maxwell’s own explanation, that he inserted citation “placeholders” meant to be corrected before filing, became the basis for a finding of deliberate fabrication: “Plaintiff could have simply inserted a generic placeholder, such as ‘FIND CASE’ or ‘CITE.’ He did not do that.”

The order applies Rule 11 to an unrepresented party directly, quoting Rule 11(b)(2)‘s text that “an attorney or unrepresented party certifies” that legal contentions are warranted by existing law. It measured the sanction by deterrence rather than compensation, declined the defendants’ per-instance formula, and took account of Maxwell’s indigence, settling on $1,000.

The citation sanction sits within a longer record of discovery disputes. The deposition-cost award in the same order stems from an earlier finding that Maxwell refused to answer questions at his first deposition, and motions to reconsider the March 26, 2026 order that set up both proceedings were pending when this order issued. For practitioners, the order shows a court inferring intent to deceive from the form of the fabricated citations themselves, without deciding how they were drafted.

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 drafting practices for unresolved citations. Here the court said the plaintiff 'could have simply inserted a generic placeholder, such as "FIND CASE" or "CITE"' rather than fabricating case details formatted as real citations, which it found showed intent to deceive.
  • Verify that quoted language appears in the cited source. The court treated paraphrases presented as direct quotations as part of the sanctionable conduct.
  • Consider that the court here did not need to resolve whether AI was used; it held the conduct sanctionable under Rule 11 either way.

Sources

Primary sources

Unverified claims:
  • AI use is not admitted or found as a fact: the court agreed the authority 'contains the hallmarks of generative AI hallucinations' but held the conduct sanctionable 'whether or not Plaintiff used AI.' No tool is identified.
  • Status of the two motions to reconsider the March 26, 2026 order (ECF Nos. 173 and 176) after May 26, 2026 was not checked.