Sanction teardown · Tribunal Superior de Bogotá, Colombia · 2025-12-02
Fiscalía General de la Nación v. Luis Alberto Contreras Fandinoe
What happened
In Tribunal Superior de Bogotá, Colombia, 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)First-instance judgment attributed a passage to 'SP414-2023, rad. 61005' about congruence; the appellate court found the cited content does not correspond to that decision.
-
Fabricated (Doctrinal Work)First-instance opinion referenced doctrinal work 'Derecho Procesal Penal Colombiano' by Fernando Velásquez Velásquez, which the appellate court found to be nonexistent.
-
Fabricated (Case Law)Decision cited multiple non-existent Corte Suprema decisions (said to support exclusion of CD loss); those citations do not exist in the repository.
-
Misrepresented (Case Law)The judgment quoted and paraphrased Constitutional Court decision C-591 de 2005 in a way the appellate court held was not supported by the original text.
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
Judgment quashed and remanded; copies sent to disciplinary commission.
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/1161/Colombia_14.pdf, via Damien Charlotin's public AI Hallucination Cases Database (CC0).