Sanction teardown · CA San José, Costa Rica · 2026-01-29
Resolución Nº 00152-2026, Expediente 19-000531-0275-PE
What happened
In CA San José, Costa Rica, a filing relied on ChatGPT to help draft legal argument. The court identified the following problems with the citations in that filing:
-
Misrepresented (Legal Norm)The brief cited article 36 of the Costa Rican Code of Criminal Procedure as authority for assessing changes in a child victim's account, but the court stated that article 36 regulates conciliation.
-
Misrepresented (Legal Norm)The brief cited article 175 of the Code of Criminal Procedure as the legal basis for principles of internal coherence, external coherence, and persistence of testimony. The court stated that those principles are not expressed in that provision or in Costa Rican legislation using those terms; article 175 concerns defective procedural acts.
-
Misrepresented (Legal Norm)The brief cited article 8 of the Code of Criminal Procedure as governing correlation between procedural documents. The court stated that article 8 concerns decisions by collegiate tribunals, not procedural correlation.
-
Misrepresented (Legal Norm)The brief invoked article 50 of the Code on Children and Adolescents to challenge the proportionality of the sentence. The court stated that this provision has no relation to sentencing proportionality.
Which AI tool
ChatGPT. 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/3273/Tribunal_de_Apelaci%C3%B3n_de_Sentencia_Penal_II_Circuito_Judicial_de_San_Jos%C3%A9_00152-2026.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).