Sanction teardown · Tennessee, USA · 2025-11-21
Shields v. First Financial
What happened
In Tennessee, 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:
-
IssueFabricated: Case Law | Doe v. Roe, No. M2018-01746-COA-R3-CV (Tenn. Ct. App. 2019) in Plaintiffs Motion for Limited Discovery Pg. 5. There is no Doe v. Roe case in the Tennessee Court of Appeals in 2019 and No. M2018-01746-COA-R3-CV returns Jones v. Jones, No. M201801746COAR3CV, 2019 WL 1036077 (Tenn. Ct. App. Mar. 5, 2019).
-
Fabricated (Case Law)Nandigam Neurology, PLC v. Beavers, No. M2021-01280-COA-R9-CV, 2022 WLE 1512346, at *5 (Tenn. Ct. App. May 13, 2022) in Plaintiffs Response to Defendant’s Motion to Compel Pg. 2. There is a Nandigam Neurology, PLC v. Beavers, 639 S.W.3d 651 (Tenn. Ct. App. 2021), but neither 2022 WL 1512346 nor No. M2021-01280-COA-R9-CV return any case in Tennessee.
-
IssueFabricated: Legal Norm
-
Fabricated (Case Law)Doe v. Roe, 477 S.W.3d 321, 326 (Tenn. Ct. App. 2015) in /<7 Pg. 4. There is no Doe v. Roe in the Tennessee Court of Appeals in 2015 and the reporter number corresponds to Ford v. State, 477 S.W.3d 321 (Tex. Crim. App. 2015).
-
Fabricated (Case Law)Schultz v. Davis, 495 S.W.3d 821, 827 (Tenn. 2016) in Plaintiffs Motion to Amend Pg. 4. There is no Schultz v. Davis case in 2016 and the reporter number corresponds to State v. Dixon, 495 S.W.3d 812 (Mo. Ct. App. 2016).
-
Fabricated (Case Law)Vincent v. Yelverton, 194 F. App'x 303, 307 (6th Cir. 2006) in Id Pg. 4. There is no Vincent v. Yelverton in the 6th Circuit in 2006 and the reporter number corresponds to United States v. Younes, 194 F. App'x 302 (6th Cir. 2006)
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/1435/Raymond_Alan_Shields_v._First_Financial_USA_21_November_2025.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).