Sanction teardown · Tribunal Supremo, Spain · date not published
BFF FINANCE IBERIA SAU v. Junta de Comunidades de Castilla-La Mancha
What happened
In Tribunal Supremo, Spain, 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)Passages reproduced in the party's written opposition, presented as quotations from Tribunal Supremo decisions, did not match the texts in the CENDOJ database upon judicial comparison.
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
Not specified in source record.
Additional detail
The respondent's written opposition included passages quoted as from Tribunal Supremo resolutions. The Court ordered the respondent to identify the cited sentences and their database references, then compared the quoted passages to the full texts in the CENDOJ database and found that the reproduced quotations did not coincide literally or substantially with the actual decisions. The Court therefore treated those citations as misrepresentations/false quotations and did not rely on them; no sanction or monetary penalty was imposed in relation to the misquoting.
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: Damien Charlotin's public AI Hallucination Cases Database (CC0).