This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Two immigration attorneys were sanctioned by the 9th U.S. Circuit Court of Appeals for submitting legal briefs containing fabricated cases generated by AI. The court found that the attorneys failed to verify their filings and violated their duty of candor by attempting to attribute the errors to typographical mistakes. As a result, the attorneys were fined and suspended from practicing before the court for six months.
California immigration attorneys Mike Singh Sethi and William Rounds were reportedly sanctioned in LNU v. Blanche after Ninth Circuit filings included purportedly fabricated and materially inaccurate legal authorities. The court said unauthorized generative AI use by unlicensed brief writers was probable and found that the attorneys failed to verify the filings and repeatedly lacked candor about the errors' source.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI systems that inadvertently generate or spread incorrect or deceptive information, which can lead to inaccurate beliefs in users and undermine their autonomy. Humans that make decisions based on false beliefs can experience physical, emotional or material harms
Also classified as
The attorneys overrelied on the AI system by failing to verify the generated citations before submitting them in a critical legal proceeding.
The AI system generated fabricated legal citations (hallucinations) which were presented as factual in a court brief, directly falling under the generation of false or misleading information.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The incident was initiated by the AI system generating fabricated legal citations (hallucinations) during post-deployment use, which were then unintentionally submitted by the attorneys.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
Two immigration attorneys were sanctioned by the 9th U.S. Circuit Court of Appeals for submitting briefs containing AI-generated fake cases. This incident represents a minor, localized disruption to federal judicial operations that was successfully mitigated using standard legal disciplinary procedures, posing negligible overall threat to national security.
The AI system acted as a writing assistant to help the attorneys draft legal briefs, augmenting their capabilities but requiring human oversight and verification which the attorneys failed to provide.
Limited Risk: The incident involves generative AI and chatbots producing content, which carries transparency obligations such as notifying the court when AI-generated content is used.
The AI system was used to assist in drafting legal briefs and generating case citations, which is a writing assistant function involving technical text generation.
The report from Reuters provides clear, factual details about the court's ruling, the names of the sanctioned attorneys, the specific penalties, and the nature of the AI-generated errors. There is no conflicting information regarding the core events.
None
Technology
Generative AI Hallucinations
The AI program fabricated nonexistent legal cases and citations.
Human Factors
Failure to Vet AI Output
Attorneys failed to check the accuracy of the AI-generated citations.
Lack of Candor
Attorneys falsely blamed errors on typographical mistakes to cover up AI use.
Subtle Subterfuge
An attorney corrected filings without disclosing the cases were fabricated.
Process and Methods
Inadequate Verification Process
No process was used to cross-reference AI-generated citations with real cases.
Regulatory Environment
Evolving Legal Standards for AI
Courts are actively defining rules and sanctions for AI-related errors.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
