Sanction teardown · W.D. Oklahoma, USA · 2026-04-07
Blackburn v. United States
What happened
In W.D. Oklahoma, 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:
-
Misrepresented (Case Law)Plaintiff cited Davoll, 194 F.3d 1116, 1138 for the proposition that Fed. R. Civ. P. 26(a)(2)(C) limits treating-physician opinion testimony as claimed; court found the pinpointed pages discuss lay-witness scope, not non-retained expert testimony — mischaracterization.
-
Misrepresented (Case Law)Plaintiff cited Muscogee (Creek) Nation, 611 F.3d 1222, 1237 as holding treating physicians may not give opinions formed in anticipation of litigation; court found that opinion does not discuss treating-physician testimony — misrepresentation.
-
Fabricated (Case Law)Plaintiff cited 'Hall v. United States, 2018 WL 1620923 (D. Colo. Apr. 3, 2018)' to support a proposition, but the provided Westlaw identifier directed to Poppino v. Berryhill, an unrelated decision that does not support the quote — a miscitation/fabricated support.
-
Misrepresented (Case Law)Plaintiff cited Davoll, 194 F.3d 1116, 1138 (10th Cir. 1999) for the proposition that treating-physician opinions are limited under Fed. R. Civ. P. 26(a)(2)(C); court noted the cited pages address a treating physician as a lay witness and do not support Plaintiff's argument.
-
Misrepresented (Case Law)Plaintiff cited Muscogee (Creek) Nation, 611 F.3d 1222, 1237 (10th Cir. 2010) as holding treating physicians may not testify to opinions developed in anticipation of litigation; court found that opinion does not discuss treating physicians.
-
Misrepresented (Case Law)Plaintiff cited Hall v. United States, 2018 WL 1620923 (D. Colo. Apr. 3, 2018) for a proposition about VA doctors; court observed the Westlaw identifier led to Poppino v. Berryhill (D.N.M.), which is unrelated and does not support the quoted proposition.
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/1965/Blackburn_v_m5HCIWT._USA_USA_7_April_2026.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).