Skip to content
Submit incident
AII-1001-07743
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.Supporting sources are pending verification.

Colombian lawyer fined for submitting AI-generated false legal citations

A Colombian lawyer was fined 26 million pesos by the Supreme Court of Justice after submitting a court appeal containing fabricated legal citations and precedents generated by an AI language model. The court ruled that lawyers have an indelegable duty to verify source authenticity before filing documents, even when using AI tools.

16 February 2026 · Jorge Hernán Zapata Vargas · Reliability and hallucination
16 February 2026
AII-1001-07743
Quick facts
Category
Reliability and hallucination
Country
Colombia
Date
16 February 2026
Drafted by
Raidu Incident Detection

What happened

Lawyer Jorge Hernán Zapata Vargas submitted an extraordinary appeal of revision to the Tribunal Superior de Villavicencio containing normas and sentencias that did not exist. The Corte Suprema de Justicia de Colombia certified that none of the cited precedents appeared in official jurisprudential systems. The lawyer acknowledged that the inaccuracies resulted "possibly from an Artificial Intelligence program that is used in order to expedite responses and procedures."

Reported impact

The Corte Suprema de Justicia imposed a fine of 26 million Colombian pesos on the lawyer for presenting judicial documents containing non-existent legal citations and precedents.

Index analysis

The court found that the lawyer committed "temeridad procesal" (procedural recklessness) by invoking miscited articles and non-existent precedents. The decision flagged the risk of introducing "pseudo-derecho" (pseudo-law)—non-existent content appearing as real jurisprudence—into judicial processes, and noted that generative language models can produce such false information.

Governance controls

The Corte Suprema de Justicia determined that lawyers have an indelegable professional duty to verify the authenticity and veracity of sources before submitting them to a judicial process, even when using technological tools. The ruling emphasized that this obligation cannot be delegated, including to AI systems.

Classification

Category
Reliability and hallucination
Industry
Government, security and defence
Country
Colombia
Harm status
Actual harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Economic or property, Reputational
Failure modes
Fabricated citation, Hallucination, Overreliance
Consequences
Fine or settlement
AI systems
Unidentified generative AI language model

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident