The Citations Were Fabricated, the AI Was Unauthorized, and the Attorneys Claimed Not to Know

What happened

In LNU v. Blanche, a case before the Ninth Circuit, California immigration attorneys Mike Singh Sethi and William Rounds were sanctioned after their filings contained legal authorities the court described as purportedly fabricated and materially inaccurate. The filings were not minor procedural paperwork. They were briefs submitted to a federal appellate court in an immigration matter, where the quality of legal argument can determine whether a person remains in the country or gets deported.

The court concluded that unauthorized use of generative AI was the probable cause of the fabricated citations. According to the findings, the briefs were likely written at least in part by unlicensed individuals who used AI tools to produce legal research and argument. Outsourcing federal appellate work to unvetted writers who rely on unverified AI output is its own serious problem, separate from the fabrications themselves. Neither attorney had apparently reviewed the work before signing the filings.

What made the sanctions sharper was how the attorneys responded once the errors were flagged. The court found that Sethi and Rounds repeatedly lacked candor about where the mistakes had come from. Candor to the tribunal is not a courtesy in federal practice; it is a professional obligation. Attorneys who submit fabricated citations face a serious problem on their own; attorneys who then obscure the source of those fabrications face a considerably worse one. The court treated that evasion as an aggravating factor rather than a mitigating one.

The immigration context matters. Clients whose cases appear before the Ninth Circuit on immigration grounds are often in the middle of removal proceedings or appeals of asylum denials. When an attorney submits a brief built on invented case citations, the court cannot rely on it and the client's case may be left without substantive argument. A fabricated legal authority cited in support of a stay or a reversal is a filing that could not have done its job even if it had gone undetected.

This case points to a gap the legal profession has not yet closed. Courts have no systematic way to know whether a brief was produced with AI assistance, whether a licensed attorney actually reviewed it, or whether the cited authorities existed before filing. The accountability structure rests entirely on attorney certification, which Sethi and Rounds undermined both by failing to verify the work and by being less than candid when questioned about it. A provable record of what a system produced, who reviewed it, and what verification occurred before submission would make that certification mean something beyond a signature.

Reported impact

Affected parties

Not publicly disclosed

Harm type

Not publicly disclosed

Scale

Not publicly disclosed

Financial impact

Not publicly disclosed

Regulatory action

Not publicly disclosed

Classification

Organization

Not publicly disclosed

AI system

Not publicly disclosed

Industry

Not publicly disclosed

Country

Not publicly disclosed

Provider

Not publicly disclosed

Incident type

Not publicly disclosed

Relevant governance controls

Governance control mapping is not available for this record.

• No controls mappedNot publicly disclosed

Control mapping is analytical. It does not state that any control would have prevented the incident.

Sources and evidence

This record was researched and written by the Index. The event is also catalogued in the following database, which is listed for cross-reference.