Sanction teardown · D. Minnesota, USA · 2026-09-30
United States v. Charles Randall Sorensen
What happened
In D. Minnesota, USA, a filing relied on an unnamed/unconfirmed AI tool to help draft legal argument. The court identified the following problems with the citations in that filing:
-
False Quotes (Case Law)Sorensen quoted nonexistent language and misrepresented the holding of United States v. Chalupnik.
-
False Quotes (Case Law)Sorensen quoted nonexistent language purportedly from United States v. Trotter.
-
False Quotes (Case Law)Sorensen quoted nonexistent language and misrepresented the holding of United States v. Jaramillo.
-
False Quotes (Case Law)Sorensen quoted nonexistent language and misrepresented the holding of United States v. Arutunoff.
-
False Quotes (Legal Norm)Sorensen quoted nonexistent language purportedly from 18 U.S.C. §§ 3612(c) and (d), and cited nonexistent installment language purportedly from 18 U.S.C. § 3664(k).
-
Misrepresented (Case Law)Sorensen quoted nonexistent language and misrepresented the holding of United States v. Hicks, a firearm-possession case that did not discuss restitution or taxes. The Court rejected its use to support enjoining IRS garnishment.
-
False Quotes (Case Law)Sorensen submitted fictitious quotations purportedly from United States v. Rodd and United States v. Loggins. The Court described the quotations as fabricated.
-
Fabricated (Case Law)Sorensen cited a nonexistent case as authority concerning IRS garnishment and restitution. The Court identified the case as fictitious.
-
False Quotes (Legal Norm)Sorensen quoted nonexistent language purportedly from the District of Minnesota’s local rule concerning waiver. The Court identified the quotation as fake.
-
False Quotes (Legal Norm)Sorensen quoted nonexistent language purportedly from the federal restitution statute concerning payment or collection.
-
False Quotes (Case Law)Sorensen quoted nonexistent language purportedly from United States v. Booth. The Court characterized the filing as containing fake quotations and irrelevant or fictitious authority.
-
False Quotes (Legal Norm)Sorensen quoted nonexistent language concerning “injustice” purportedly from Federal Rule of Criminal Procedure 57(b).
-
False Quotes (Case Law)Sorensen quoted nonexistent language and misrepresented the holding of United States v. Malewicka.
-
False Quotes (Case Law)Sorensen quoted nonexistent language purportedly from Miller v. United States.
-
False Quotes (Case Law)Sorensen quoted nonexistent language purportedly from Degen v. United States.
-
False Quotes (Case Law)Sorensen quoted nonexistent language purportedly from United States v. ITT Continental Baking Co.
-
False Quotes (Exhibits & Submissions)Sorensen quoted nonexistent language about restitution and coordination with criminal judgments purportedly from an IRS manual.
-
False Quotes (Doctrinal Work)Sorensen quoted nonexistent language purportedly from a United States Sentencing Commission report.
-
False Quotes (Case Law)Sorensen quoted nonexistent language purportedly from Gelco Corp. v. Coniston Partners.
-
False Quotes (Case Law)Sorensen quoted nonexistent language purportedly from Commodity Futures Trading Commission v. Co Petro Marketing Group.
-
False Quotes (Case Law)Sorensen quoted nonexistent language purportedly from Amoco Production Co. v. Village of Gambell.
Which AI tool
an unnamed/unconfirmed AI tool. Note: Charlotin's public database records tool attribution only where a court order, brief, or reporting on the matter states it explicitly; "unidentified" or "implied" means the record indicates AI use but does not name a specific product — we do not guess.
Outcome
Warning
How Citation Safe would have caught this
Citation Safe runs three deterministic layers before a brief is filed: (1) does the citation exist against CourtListener's database of published opinions, (2) if quoted, does that exact language appear in the source, (3) does the cited case actually support the proposition it is cited for. Fabricated case citations fail Layer 1. Fabricated or misattributed quotations fail Layer 2 even when the underlying case is real. Misrepresented holdings — a real case cited for a proposition it does not support — are the target of Layer 3. None of these checks involve asking another language model whether the citation looks right; they are lookups and text-matches against the actual source, which is why a hallucinated citation has to survive a direct lookup against the authoritative source — not another model's opinion — to earn a VERIFIED stamp; our measured false-verify rate is published live at /quality.
Check a brief before you file it → · See our live false-verify rate
Source: https://www.damiencharlotin.com/documents/3167/USA_v._Sorensen_USA_30-09-2026.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).